diff --git a/api/v1/meta.json b/api/v1/meta.json index f5951f9..007faf7 100644 --- a/api/v1/meta.json +++ b/api/v1/meta.json @@ -1,10 +1,10 @@ { "dataset_name": "AICostBudget AI API Pricing Dataset", "dataset_version": "1.0.0", - "generated_at": "2026-07-09T18:36:17Z", + "generated_at": "2026-07-27T00:00:00Z", "homepage": "https://aicostbudget.com", - "last_verified_at": "2026-07-10T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_count": 21, - "official_source_count": 9, + "official_source_count": 10, "provider_count": 7 } diff --git a/api/v1/models/anthropic/claude-haiku-4.5.json b/api/v1/models/anthropic/claude-haiku-4.5.json index 870db72..c201ead 100644 --- a/api/v1/models/anthropic/claude-haiku-4.5.json +++ b/api/v1/models/anthropic/claude-haiku-4.5.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Claude Haiku 4.5", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Claude Haiku", "model_id": "claude-haiku-4.5", "notes": "cache_write stores the 5-minute write rate.", diff --git a/api/v1/models/anthropic/claude-opus-4.8.json b/api/v1/models/anthropic/claude-opus-4.8.json index 2ee11c5..d5e2db9 100644 --- a/api/v1/models/anthropic/claude-opus-4.8.json +++ b/api/v1/models/anthropic/claude-opus-4.8.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Claude Opus 4.8", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Claude Opus", "model_id": "claude-opus-4.8", "notes": "Anthropic table columns are input, 5-minute cache write, 1-hour cache write, cache read, and output. cache_write stores the 5-minute write rate.", diff --git a/api/v1/models/anthropic/claude-sonnet-4.6.json b/api/v1/models/anthropic/claude-sonnet-4.6.json index 8e566c8..e9a5590 100644 --- a/api/v1/models/anthropic/claude-sonnet-4.6.json +++ b/api/v1/models/anthropic/claude-sonnet-4.6.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Claude Sonnet 4.6", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Claude Sonnet", "model_id": "claude-sonnet-4.6", "notes": "cache_write stores the 5-minute write rate.", diff --git a/api/v1/models/anthropic/claude-sonnet-5-intro.json b/api/v1/models/anthropic/claude-sonnet-5-intro.json index a24f635..af4499a 100644 --- a/api/v1/models/anthropic/claude-sonnet-5-intro.json +++ b/api/v1/models/anthropic/claude-sonnet-5-intro.json @@ -1,12 +1,12 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Claude Sonnet 5 introductory pricing", - "effective_from": "2026-07-05", - "last_verified_at": "2026-07-05T00:00:00Z", + "effective_from": "2026-06-30", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Claude Sonnet", "model_id": "claude-sonnet-5-intro", - "notes": "Introductory pricing is listed by Anthropic as effective through 2026-08-31; standard pricing starts 2026-09-01.", - "official_source_url": "https://docs.anthropic.com/en/docs/about-claude/pricing#claude-sonnet-5-introductory-pricing", + "notes": "Introductory pricing applies through 2026-08-31 and standard pricing starts 2026-09-01. V1 must not switch early to the standard $3/M input and $15/M output rates.", + "official_source_url": "https://www.anthropic.com/news/claude-sonnet-5", "pricing": { "batch_input": 1.0, "batch_output": 5.0, diff --git a/api/v1/models/cohere/aya-expanse-32b.json b/api/v1/models/cohere/aya-expanse-32b.json index f414dbd..f8ba278 100644 --- a/api/v1/models/cohere/aya-expanse-32b.json +++ b/api/v1/models/cohere/aya-expanse-32b.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Aya Expanse 32B", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Aya", "model_id": "aya-expanse-32b", "notes": "Cohere official pricing page states Aya Expanse 8B and 32B API pricing is $0.50/1M input and $1.50/1M output.", diff --git a/api/v1/models/cohere/command-a-plus.json b/api/v1/models/cohere/command-a-plus.json index 3e70177..0d4f45b 100644 --- a/api/v1/models/cohere/command-a-plus.json +++ b/api/v1/models/cohere/command-a-plus.json @@ -1,12 +1,12 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Command A+", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Command", "model_id": "command-a-plus", - "notes": "Cohere official pricing page lists Command A+ but the extracted public page did not expose a confirmed token price. Values are null rather than guessed.", - "official_source_url": "https://cohere.com/pricing", + "notes": "Cohere's official Command A+ page states that access is free until applicable rate limits are reached and that production use is available through Model Vault. Another official Cohere page shows per-token pricing for the same model ID, so V1 keeps token pricing null until the official documentation is reconciled.", + "official_source_url": "https://docs.cohere.com/docs/command-a-plus", "pricing": { "batch_input": null, "batch_output": null, diff --git a/api/v1/models/cohere/command-r-plus-08-2024.json b/api/v1/models/cohere/command-r-plus-08-2024.json index f9e6f83..d10294b 100644 --- a/api/v1/models/cohere/command-r-plus-08-2024.json +++ b/api/v1/models/cohere/command-r-plus-08-2024.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Command R+ 08-2024", "effective_from": "2024-08-01", - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Command R", "model_id": "command-r-plus-08-2024", "notes": "Cohere official pricing page lists Command R+ 08-2024 at $2.50/1M input tokens and $10.00/1M output tokens.", diff --git a/api/v1/models/deepseek/deepseek-v4-flash.json b/api/v1/models/deepseek/deepseek-v4-flash.json index a85837f..52f31ad 100644 --- a/api/v1/models/deepseek/deepseek-v4-flash.json +++ b/api/v1/models/deepseek/deepseek-v4-flash.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "DeepSeek V4 Flash", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "DeepSeek V4", "model_id": "deepseek-v4-flash", "notes": "DeepSeek labels input cache miss as input price and cache hit as cached_input.", diff --git a/api/v1/models/deepseek/deepseek-v4-pro.json b/api/v1/models/deepseek/deepseek-v4-pro.json index e5f98bc..1d6a1f6 100644 --- a/api/v1/models/deepseek/deepseek-v4-pro.json +++ b/api/v1/models/deepseek/deepseek-v4-pro.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "DeepSeek V4 Pro", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "DeepSeek V4", "model_id": "deepseek-v4-pro", "notes": "DeepSeek labels input cache miss as input price and cache hit as cached_input.", diff --git a/api/v1/models/google-gemini/gemini-2.5-flash.json b/api/v1/models/google-gemini/gemini-2.5-flash.json index 69bbec5..91bea1e 100644 --- a/api/v1/models/google-gemini/gemini-2.5-flash.json +++ b/api/v1/models/google-gemini/gemini-2.5-flash.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Gemini 2.5 Flash", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Gemini 2.5", "model_id": "gemini-2.5-flash", "notes": "Rates are for text/image/video input; Google lists separate audio rates in the official pricing table.", diff --git a/api/v1/models/google-gemini/gemini-2.5-pro.json b/api/v1/models/google-gemini/gemini-2.5-pro.json index fb63bed..32997b2 100644 --- a/api/v1/models/google-gemini/gemini-2.5-pro.json +++ b/api/v1/models/google-gemini/gemini-2.5-pro.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Gemini 2.5 Pro", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Gemini 2.5", "model_id": "gemini-2.5-pro", "notes": "Rates are for prompts up to 200k tokens. Google lists higher rates above 200k tokens in the official pricing table.", diff --git a/api/v1/models/mistral-ai/mistral-large.json b/api/v1/models/mistral-ai/mistral-large.json index 4b20e7b..109a096 100644 --- a/api/v1/models/mistral-ai/mistral-large.json +++ b/api/v1/models/mistral-ai/mistral-large.json @@ -1,20 +1,20 @@ { - "accessed_at": "2026-07-05T00:00:00Z", - "display_name": "Mistral Large", + "accessed_at": "2026-07-27T00:00:00Z", + "display_name": "Mistral Large 3", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Mistral Large", "model_id": "mistral-large", - "notes": "Mistral official pricing FAQ states Mistral Large costs $2/M input and $6/M output and batch processing gets a 50% discount.", - "official_source_url": "https://mistral.ai/pricing", + "notes": "mistral-large-latest currently maps to Mistral Large 3. Standard price is $0.50/M input and $1.50/M output. Cached input receives a 90% discount. Batch processing receives a 50% discount.", + "official_source_url": "https://mistral.ai/pricing/api/", "pricing": { - "batch_input": 1.0, - "batch_output": 3.0, + "batch_input": 0.25, + "batch_output": 0.75, "cache_write": null, - "cached_input": null, + "cached_input": 0.05, "currency": "USD", - "input": 2.0, - "output": 6.0, + "input": 0.5, + "output": 1.5, "unit": "1M tokens" }, "provider_id": "mistral-ai", diff --git a/api/v1/models/openai/gpt-4.1.json b/api/v1/models/openai/gpt-4.1.json index c2da567..6bc6e29 100644 --- a/api/v1/models/openai/gpt-4.1.json +++ b/api/v1/models/openai/gpt-4.1.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-4.1", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-4.1", "model_id": "gpt-4.1", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", diff --git a/api/v1/models/openai/gpt-5.4-mini.json b/api/v1/models/openai/gpt-5.4-mini.json index a46ae5f..0207b01 100644 --- a/api/v1/models/openai/gpt-5.4-mini.json +++ b/api/v1/models/openai/gpt-5.4-mini.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5.4 mini", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5.4", "model_id": "gpt-5.4-mini", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", diff --git a/api/v1/models/openai/gpt-5.5.json b/api/v1/models/openai/gpt-5.5.json index 39e1e09..07f4d09 100644 --- a/api/v1/models/openai/gpt-5.5.json +++ b/api/v1/models/openai/gpt-5.5.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5.5", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5.5", "model_id": "gpt-5.5", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens. Batch prices are 50% of the listed standard price where shown.", diff --git a/api/v1/models/openai/gpt-5.6-luna.json b/api/v1/models/openai/gpt-5.6-luna.json index 515bccf..12aecbb 100644 --- a/api/v1/models/openai/gpt-5.6-luna.json +++ b/api/v1/models/openai/gpt-5.6-luna.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-10T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5.6 Luna", "effective_from": null, - "last_verified_at": "2026-07-10T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5.6", "model_id": "gpt-5.6-luna", "notes": "OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices.", diff --git a/api/v1/models/openai/gpt-5.6-sol.json b/api/v1/models/openai/gpt-5.6-sol.json index 8cb9bfa..8fe085e 100644 --- a/api/v1/models/openai/gpt-5.6-sol.json +++ b/api/v1/models/openai/gpt-5.6-sol.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-10T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5.6 Sol", "effective_from": null, - "last_verified_at": "2026-07-10T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5.6", "model_id": "gpt-5.6-sol", "notes": "OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices.", diff --git a/api/v1/models/openai/gpt-5.6-terra.json b/api/v1/models/openai/gpt-5.6-terra.json index 1c0fa46..3531a31 100644 --- a/api/v1/models/openai/gpt-5.6-terra.json +++ b/api/v1/models/openai/gpt-5.6-terra.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-10T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5.6 Terra", "effective_from": null, - "last_verified_at": "2026-07-10T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5.6", "model_id": "gpt-5.6-terra", "notes": "OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices.", diff --git a/api/v1/models/openai/gpt-5.json b/api/v1/models/openai/gpt-5.json index 76badf4..1ba3653 100644 --- a/api/v1/models/openai/gpt-5.json +++ b/api/v1/models/openai/gpt-5.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5", "model_id": "gpt-5", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", diff --git a/api/v1/models/openai/o3.json b/api/v1/models/openai/o3.json index a9aad76..4f9c4e3 100644 --- a/api/v1/models/openai/o3.json +++ b/api/v1/models/openai/o3.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "o3", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "o-series", "model_id": "o3", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", diff --git a/api/v1/models/xai/grok-4.3.json b/api/v1/models/xai/grok-4.3.json index f14c96d..f4d16b8 100644 --- a/api/v1/models/xai/grok-4.3.json +++ b/api/v1/models/xai/grok-4.3.json @@ -1,17 +1,17 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Grok 4.3", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Grok 4", "model_id": "grok-4.3", - "notes": "xAI official model page lists Grok 4.3 input and output rates per 1M tokens. Cached and batch rates were not shown in the extracted official model card.", - "official_source_url": "https://docs.x.ai/developers/models", + "notes": "V1 stores short-context pricing. The long-context threshold is 200K tokens. Grok 4.3 batch pricing receives a 20% discount. Long-context prices are not stored in the current short-context fields.", + "official_source_url": "https://docs.x.ai/developers/pricing", "pricing": { - "batch_input": null, - "batch_output": null, + "batch_input": 1.0, + "batch_output": 2.0, "cache_write": null, - "cached_input": null, + "cached_input": 0.2, "currency": "USD", "input": 1.25, "output": 2.5, diff --git a/api/v1/prices.csv b/api/v1/prices.csv index e88a766..47b95df 100644 --- a/api/v1/prices.csv +++ b/api/v1/prices.csv @@ -1,22 +1,22 @@ provider_id,model_id,display_name,model_family,status,currency,unit,input,output,cached_input,cache_write,batch_input,batch_output,official_source_url,accessed_at,last_verified_at,effective_from,notes -anthropic,claude-haiku-4.5,Claude Haiku 4.5,Claude Haiku,active,USD,1M tokens,1.0,5.0,0.1,1.25,0.5,2.5,https://docs.anthropic.com/en/docs/about-claude/pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,,cache_write stores the 5-minute write rate. -anthropic,claude-opus-4.8,Claude Opus 4.8,Claude Opus,active,USD,1M tokens,5.0,25.0,0.5,6.25,2.5,12.5,https://docs.anthropic.com/en/docs/about-claude/pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,,"Anthropic table columns are input, 5-minute cache write, 1-hour cache write, cache read, and output. cache_write stores the 5-minute write rate." -anthropic,claude-sonnet-4.6,Claude Sonnet 4.6,Claude Sonnet,active,USD,1M tokens,3.0,15.0,0.3,3.75,1.5,7.5,https://docs.anthropic.com/en/docs/about-claude/pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,,cache_write stores the 5-minute write rate. -anthropic,claude-sonnet-5-intro,Claude Sonnet 5 introductory pricing,Claude Sonnet,active,USD,1M tokens,2.0,10.0,0.2,2.5,1.0,5.0,https://docs.anthropic.com/en/docs/about-claude/pricing#claude-sonnet-5-introductory-pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,2026-07-05,Introductory pricing is listed by Anthropic as effective through 2026-08-31; standard pricing starts 2026-09-01. -cohere,aya-expanse-32b,Aya Expanse 32B,Aya,active,USD,1M tokens,0.5,1.5,,,,,https://cohere.com/pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,,Cohere official pricing page states Aya Expanse 8B and 32B API pricing is $0.50/1M input and $1.50/1M output. -cohere,command-a-plus,Command A+,Command,active,USD,1M tokens,,,,,,,https://cohere.com/pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,,Cohere official pricing page lists Command A+ but the extracted public page did not expose a confirmed token price. Values are null rather than guessed. -cohere,command-r-plus-08-2024,Command R+ 08-2024,Command R,active,USD,1M tokens,2.5,10.0,,,,,https://cohere.com/pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,2024-08-01,Cohere official pricing page lists Command R+ 08-2024 at $2.50/1M input tokens and $10.00/1M output tokens. -deepseek,deepseek-v4-flash,DeepSeek V4 Flash,DeepSeek V4,active,USD,1M tokens,0.14,0.28,0.0028,,,,https://api-docs.deepseek.com/quick_start/pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,,DeepSeek labels input cache miss as input price and cache hit as cached_input. -deepseek,deepseek-v4-pro,DeepSeek V4 Pro,DeepSeek V4,active,USD,1M tokens,0.435,0.87,0.003625,,,,https://api-docs.deepseek.com/quick_start/pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,,DeepSeek labels input cache miss as input price and cache hit as cached_input. -google-gemini,gemini-2.5-flash,Gemini 2.5 Flash,Gemini 2.5,active,USD,1M tokens,0.3,2.5,0.03,,0.15,1.25,https://ai.google.dev/gemini-api/docs/pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,,Rates are for text/image/video input; Google lists separate audio rates in the official pricing table. -google-gemini,gemini-2.5-pro,Gemini 2.5 Pro,Gemini 2.5,active,USD,1M tokens,1.25,10.0,0.125,,0.625,5.0,https://ai.google.dev/gemini-api/docs/pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,,Rates are for prompts up to 200k tokens. Google lists higher rates above 200k tokens in the official pricing table. -mistral-ai,mistral-large,Mistral Large,Mistral Large,active,USD,1M tokens,2.0,6.0,,,1.0,3.0,https://mistral.ai/pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,,Mistral official pricing FAQ states Mistral Large costs $2/M input and $6/M output and batch processing gets a 50% discount. -openai,gpt-4.1,GPT-4.1,GPT-4.1,active,USD,1M tokens,2.0,8.0,0.5,,1.0,4.0,https://platform.openai.com/docs/pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,,OpenAI official pricing page lists standard and batch rates per 1M tokens. -openai,gpt-5,GPT-5,GPT-5,active,USD,1M tokens,1.25,10.0,0.125,,0.625,5.0,https://platform.openai.com/docs/pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,,OpenAI official pricing page lists standard and batch rates per 1M tokens. -openai,gpt-5.4-mini,GPT-5.4 mini,GPT-5.4,active,USD,1M tokens,0.75,4.5,0.075,,0.375,2.25,https://platform.openai.com/docs/pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,,OpenAI official pricing page lists standard and batch rates per 1M tokens. -openai,gpt-5.5,GPT-5.5,GPT-5.5,active,USD,1M tokens,5.0,30.0,0.5,,2.5,15.0,https://platform.openai.com/docs/pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,,OpenAI official pricing page lists standard and batch rates per 1M tokens. Batch prices are 50% of the listed standard price where shown. -openai,gpt-5.6-luna,GPT-5.6 Luna,GPT-5.6,active,USD,1M tokens,1.0,6.0,0.1,1.25,0.5,3.0,https://developers.openai.com/api/docs/pricing,2026-07-10T00:00:00Z,2026-07-10T00:00:00Z,,"OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices." -openai,gpt-5.6-sol,GPT-5.6 Sol,GPT-5.6,active,USD,1M tokens,5.0,30.0,0.5,6.25,2.5,15.0,https://developers.openai.com/api/docs/pricing,2026-07-10T00:00:00Z,2026-07-10T00:00:00Z,,"OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices." -openai,gpt-5.6-terra,GPT-5.6 Terra,GPT-5.6,active,USD,1M tokens,2.5,15.0,0.25,3.125,1.25,7.5,https://developers.openai.com/api/docs/pricing,2026-07-10T00:00:00Z,2026-07-10T00:00:00Z,,"OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices." -openai,o3,o3,o-series,active,USD,1M tokens,2.0,8.0,0.5,,1.0,4.0,https://platform.openai.com/docs/pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,,OpenAI official pricing page lists standard and batch rates per 1M tokens. -xai,grok-4.3,Grok 4.3,Grok 4,active,USD,1M tokens,1.25,2.5,,,,,https://docs.x.ai/developers/models,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,,xAI official model page lists Grok 4.3 input and output rates per 1M tokens. Cached and batch rates were not shown in the extracted official model card. +anthropic,claude-haiku-4.5,Claude Haiku 4.5,Claude Haiku,active,USD,1M tokens,1.0,5.0,0.1,1.25,0.5,2.5,https://docs.anthropic.com/en/docs/about-claude/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,cache_write stores the 5-minute write rate. +anthropic,claude-opus-4.8,Claude Opus 4.8,Claude Opus,active,USD,1M tokens,5.0,25.0,0.5,6.25,2.5,12.5,https://docs.anthropic.com/en/docs/about-claude/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,"Anthropic table columns are input, 5-minute cache write, 1-hour cache write, cache read, and output. cache_write stores the 5-minute write rate." +anthropic,claude-sonnet-4.6,Claude Sonnet 4.6,Claude Sonnet,active,USD,1M tokens,3.0,15.0,0.3,3.75,1.5,7.5,https://docs.anthropic.com/en/docs/about-claude/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,cache_write stores the 5-minute write rate. +anthropic,claude-sonnet-5-intro,Claude Sonnet 5 introductory pricing,Claude Sonnet,active,USD,1M tokens,2.0,10.0,0.2,2.5,1.0,5.0,https://www.anthropic.com/news/claude-sonnet-5,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,2026-06-30,Introductory pricing applies through 2026-08-31 and standard pricing starts 2026-09-01. V1 must not switch early to the standard $3/M input and $15/M output rates. +cohere,aya-expanse-32b,Aya Expanse 32B,Aya,active,USD,1M tokens,0.5,1.5,,,,,https://cohere.com/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,Cohere official pricing page states Aya Expanse 8B and 32B API pricing is $0.50/1M input and $1.50/1M output. +cohere,command-a-plus,Command A+,Command,active,USD,1M tokens,,,,,,,https://docs.cohere.com/docs/command-a-plus,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,"Cohere's official Command A+ page states that access is free until applicable rate limits are reached and that production use is available through Model Vault. Another official Cohere page shows per-token pricing for the same model ID, so V1 keeps token pricing null until the official documentation is reconciled." +cohere,command-r-plus-08-2024,Command R+ 08-2024,Command R,active,USD,1M tokens,2.5,10.0,,,,,https://cohere.com/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,2024-08-01,Cohere official pricing page lists Command R+ 08-2024 at $2.50/1M input tokens and $10.00/1M output tokens. +deepseek,deepseek-v4-flash,DeepSeek V4 Flash,DeepSeek V4,active,USD,1M tokens,0.14,0.28,0.0028,,,,https://api-docs.deepseek.com/quick_start/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,DeepSeek labels input cache miss as input price and cache hit as cached_input. +deepseek,deepseek-v4-pro,DeepSeek V4 Pro,DeepSeek V4,active,USD,1M tokens,0.435,0.87,0.003625,,,,https://api-docs.deepseek.com/quick_start/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,DeepSeek labels input cache miss as input price and cache hit as cached_input. +google-gemini,gemini-2.5-flash,Gemini 2.5 Flash,Gemini 2.5,active,USD,1M tokens,0.3,2.5,0.03,,0.15,1.25,https://ai.google.dev/gemini-api/docs/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,Rates are for text/image/video input; Google lists separate audio rates in the official pricing table. +google-gemini,gemini-2.5-pro,Gemini 2.5 Pro,Gemini 2.5,active,USD,1M tokens,1.25,10.0,0.125,,0.625,5.0,https://ai.google.dev/gemini-api/docs/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,Rates are for prompts up to 200k tokens. Google lists higher rates above 200k tokens in the official pricing table. +mistral-ai,mistral-large,Mistral Large 3,Mistral Large,active,USD,1M tokens,0.5,1.5,0.05,,0.25,0.75,https://mistral.ai/pricing/api/,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,mistral-large-latest currently maps to Mistral Large 3. Standard price is $0.50/M input and $1.50/M output. Cached input receives a 90% discount. Batch processing receives a 50% discount. +openai,gpt-4.1,GPT-4.1,GPT-4.1,active,USD,1M tokens,2.0,8.0,0.5,,1.0,4.0,https://platform.openai.com/docs/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,OpenAI official pricing page lists standard and batch rates per 1M tokens. +openai,gpt-5,GPT-5,GPT-5,active,USD,1M tokens,1.25,10.0,0.125,,0.625,5.0,https://platform.openai.com/docs/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,OpenAI official pricing page lists standard and batch rates per 1M tokens. +openai,gpt-5.4-mini,GPT-5.4 mini,GPT-5.4,active,USD,1M tokens,0.75,4.5,0.075,,0.375,2.25,https://platform.openai.com/docs/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,OpenAI official pricing page lists standard and batch rates per 1M tokens. +openai,gpt-5.5,GPT-5.5,GPT-5.5,active,USD,1M tokens,5.0,30.0,0.5,,2.5,15.0,https://platform.openai.com/docs/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,OpenAI official pricing page lists standard and batch rates per 1M tokens. Batch prices are 50% of the listed standard price where shown. +openai,gpt-5.6-luna,GPT-5.6 Luna,GPT-5.6,active,USD,1M tokens,1.0,6.0,0.1,1.25,0.5,3.0,https://developers.openai.com/api/docs/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,"OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices." +openai,gpt-5.6-sol,GPT-5.6 Sol,GPT-5.6,active,USD,1M tokens,5.0,30.0,0.5,6.25,2.5,15.0,https://developers.openai.com/api/docs/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,"OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices." +openai,gpt-5.6-terra,GPT-5.6 Terra,GPT-5.6,active,USD,1M tokens,2.5,15.0,0.25,3.125,1.25,7.5,https://developers.openai.com/api/docs/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,"OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices." +openai,o3,o3,o-series,active,USD,1M tokens,2.0,8.0,0.5,,1.0,4.0,https://platform.openai.com/docs/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,OpenAI official pricing page lists standard and batch rates per 1M tokens. +xai,grok-4.3,Grok 4.3,Grok 4,active,USD,1M tokens,1.25,2.5,0.2,,1.0,2.0,https://docs.x.ai/developers/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,V1 stores short-context pricing. The long-context threshold is 200K tokens. Grok 4.3 batch pricing receives a 20% discount. Long-context prices are not stored in the current short-context fields. diff --git a/api/v1/prices.json b/api/v1/prices.json index 3967948..21cf181 100644 --- a/api/v1/prices.json +++ b/api/v1/prices.json @@ -2,9 +2,9 @@ "dataset_name": "AICostBudget AI API Pricing Dataset", "dataset_version": "1.0.0", "description": "Open, machine-readable pricing data for LLM and AI APIs.", - "generated_at": "2026-07-09T18:36:17Z", + "generated_at": "2026-07-27T00:00:00Z", "homepage": "https://aicostbudget.com", - "last_verified_at": "2026-07-10T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "licenses": { "code": "MIT", "data": "CC BY 4.0" @@ -12,10 +12,10 @@ "model_count": 21, "models": [ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Claude Haiku 4.5", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Claude Haiku", "model_id": "claude-haiku-4.5", "notes": "cache_write stores the 5-minute write rate.", @@ -34,10 +34,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Claude Opus 4.8", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Claude Opus", "model_id": "claude-opus-4.8", "notes": "Anthropic table columns are input, 5-minute cache write, 1-hour cache write, cache read, and output. cache_write stores the 5-minute write rate.", @@ -56,10 +56,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Claude Sonnet 4.6", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Claude Sonnet", "model_id": "claude-sonnet-4.6", "notes": "cache_write stores the 5-minute write rate.", @@ -78,14 +78,14 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Claude Sonnet 5 introductory pricing", - "effective_from": "2026-07-05", - "last_verified_at": "2026-07-05T00:00:00Z", + "effective_from": "2026-06-30", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Claude Sonnet", "model_id": "claude-sonnet-5-intro", - "notes": "Introductory pricing is listed by Anthropic as effective through 2026-08-31; standard pricing starts 2026-09-01.", - "official_source_url": "https://docs.anthropic.com/en/docs/about-claude/pricing#claude-sonnet-5-introductory-pricing", + "notes": "Introductory pricing applies through 2026-08-31 and standard pricing starts 2026-09-01. V1 must not switch early to the standard $3/M input and $15/M output rates.", + "official_source_url": "https://www.anthropic.com/news/claude-sonnet-5", "pricing": { "batch_input": 1.0, "batch_output": 5.0, @@ -100,10 +100,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Aya Expanse 32B", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Aya", "model_id": "aya-expanse-32b", "notes": "Cohere official pricing page states Aya Expanse 8B and 32B API pricing is $0.50/1M input and $1.50/1M output.", @@ -122,14 +122,14 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Command A+", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Command", "model_id": "command-a-plus", - "notes": "Cohere official pricing page lists Command A+ but the extracted public page did not expose a confirmed token price. Values are null rather than guessed.", - "official_source_url": "https://cohere.com/pricing", + "notes": "Cohere's official Command A+ page states that access is free until applicable rate limits are reached and that production use is available through Model Vault. Another official Cohere page shows per-token pricing for the same model ID, so V1 keeps token pricing null until the official documentation is reconciled.", + "official_source_url": "https://docs.cohere.com/docs/command-a-plus", "pricing": { "batch_input": null, "batch_output": null, @@ -144,10 +144,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Command R+ 08-2024", "effective_from": "2024-08-01", - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Command R", "model_id": "command-r-plus-08-2024", "notes": "Cohere official pricing page lists Command R+ 08-2024 at $2.50/1M input tokens and $10.00/1M output tokens.", @@ -166,10 +166,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "DeepSeek V4 Flash", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "DeepSeek V4", "model_id": "deepseek-v4-flash", "notes": "DeepSeek labels input cache miss as input price and cache hit as cached_input.", @@ -188,10 +188,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "DeepSeek V4 Pro", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "DeepSeek V4", "model_id": "deepseek-v4-pro", "notes": "DeepSeek labels input cache miss as input price and cache hit as cached_input.", @@ -210,10 +210,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Gemini 2.5 Flash", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Gemini 2.5", "model_id": "gemini-2.5-flash", "notes": "Rates are for text/image/video input; Google lists separate audio rates in the official pricing table.", @@ -232,10 +232,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Gemini 2.5 Pro", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Gemini 2.5", "model_id": "gemini-2.5-pro", "notes": "Rates are for prompts up to 200k tokens. Google lists higher rates above 200k tokens in the official pricing table.", @@ -254,32 +254,32 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", - "display_name": "Mistral Large", + "accessed_at": "2026-07-27T00:00:00Z", + "display_name": "Mistral Large 3", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Mistral Large", "model_id": "mistral-large", - "notes": "Mistral official pricing FAQ states Mistral Large costs $2/M input and $6/M output and batch processing gets a 50% discount.", - "official_source_url": "https://mistral.ai/pricing", + "notes": "mistral-large-latest currently maps to Mistral Large 3. Standard price is $0.50/M input and $1.50/M output. Cached input receives a 90% discount. Batch processing receives a 50% discount.", + "official_source_url": "https://mistral.ai/pricing/api/", "pricing": { - "batch_input": 1.0, - "batch_output": 3.0, + "batch_input": 0.25, + "batch_output": 0.75, "cache_write": null, - "cached_input": null, + "cached_input": 0.05, "currency": "USD", - "input": 2.0, - "output": 6.0, + "input": 0.5, + "output": 1.5, "unit": "1M tokens" }, "provider_id": "mistral-ai", "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-4.1", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-4.1", "model_id": "gpt-4.1", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", @@ -298,10 +298,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5", "model_id": "gpt-5", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", @@ -320,10 +320,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5.4 mini", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5.4", "model_id": "gpt-5.4-mini", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", @@ -342,10 +342,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5.5", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5.5", "model_id": "gpt-5.5", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens. Batch prices are 50% of the listed standard price where shown.", @@ -364,10 +364,10 @@ "status": "active" }, { - "accessed_at": "2026-07-10T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5.6 Luna", "effective_from": null, - "last_verified_at": "2026-07-10T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5.6", "model_id": "gpt-5.6-luna", "notes": "OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices.", @@ -386,10 +386,10 @@ "status": "active" }, { - "accessed_at": "2026-07-10T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5.6 Sol", "effective_from": null, - "last_verified_at": "2026-07-10T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5.6", "model_id": "gpt-5.6-sol", "notes": "OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices.", @@ -408,10 +408,10 @@ "status": "active" }, { - "accessed_at": "2026-07-10T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5.6 Terra", "effective_from": null, - "last_verified_at": "2026-07-10T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5.6", "model_id": "gpt-5.6-terra", "notes": "OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices.", @@ -430,10 +430,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "o3", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "o-series", "model_id": "o3", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", @@ -452,19 +452,19 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Grok 4.3", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Grok 4", "model_id": "grok-4.3", - "notes": "xAI official model page lists Grok 4.3 input and output rates per 1M tokens. Cached and batch rates were not shown in the extracted official model card.", - "official_source_url": "https://docs.x.ai/developers/models", + "notes": "V1 stores short-context pricing. The long-context threshold is 200K tokens. Grok 4.3 batch pricing receives a 20% discount. Long-context prices are not stored in the current short-context fields.", + "official_source_url": "https://docs.x.ai/developers/pricing", "pricing": { - "batch_input": null, - "batch_output": null, + "batch_input": 1.0, + "batch_output": 2.0, "cache_write": null, - "cached_input": null, + "cached_input": 0.2, "currency": "USD", "input": 1.25, "output": 2.5, @@ -474,17 +474,18 @@ "status": "active" } ], - "official_source_count": 9, + "official_source_count": 10, "official_sources": [ "https://ai.google.dev/gemini-api/docs/pricing", "https://api-docs.deepseek.com/quick_start/pricing", "https://cohere.com/pricing", "https://developers.openai.com/api/docs/pricing", "https://docs.anthropic.com/en/docs/about-claude/pricing", - "https://docs.anthropic.com/en/docs/about-claude/pricing#claude-sonnet-5-introductory-pricing", - "https://docs.x.ai/developers/models", - "https://mistral.ai/pricing", - "https://platform.openai.com/docs/pricing" + "https://docs.cohere.com/docs/command-a-plus", + "https://docs.x.ai/developers/pricing", + "https://mistral.ai/pricing/api/", + "https://platform.openai.com/docs/pricing", + "https://www.anthropic.com/news/claude-sonnet-5" ], "provider_count": 7, "providers": [ diff --git a/api/v1/providers/anthropic.json b/api/v1/providers/anthropic.json index 41b2fa6..5e7be6c 100644 --- a/api/v1/providers/anthropic.json +++ b/api/v1/providers/anthropic.json @@ -3,10 +3,10 @@ "docs_url": "https://docs.anthropic.com", "models": [ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Claude Haiku 4.5", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Claude Haiku", "model_id": "claude-haiku-4.5", "notes": "cache_write stores the 5-minute write rate.", @@ -25,10 +25,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Claude Opus 4.8", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Claude Opus", "model_id": "claude-opus-4.8", "notes": "Anthropic table columns are input, 5-minute cache write, 1-hour cache write, cache read, and output. cache_write stores the 5-minute write rate.", @@ -47,10 +47,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Claude Sonnet 4.6", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Claude Sonnet", "model_id": "claude-sonnet-4.6", "notes": "cache_write stores the 5-minute write rate.", @@ -69,14 +69,14 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Claude Sonnet 5 introductory pricing", - "effective_from": "2026-07-05", - "last_verified_at": "2026-07-05T00:00:00Z", + "effective_from": "2026-06-30", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Claude Sonnet", "model_id": "claude-sonnet-5-intro", - "notes": "Introductory pricing is listed by Anthropic as effective through 2026-08-31; standard pricing starts 2026-09-01.", - "official_source_url": "https://docs.anthropic.com/en/docs/about-claude/pricing#claude-sonnet-5-introductory-pricing", + "notes": "Introductory pricing applies through 2026-08-31 and standard pricing starts 2026-09-01. V1 must not switch early to the standard $3/M input and $15/M output rates.", + "official_source_url": "https://www.anthropic.com/news/claude-sonnet-5", "pricing": { "batch_input": 1.0, "batch_output": 5.0, diff --git a/api/v1/providers/cohere.json b/api/v1/providers/cohere.json index d9b1bd9..c6cadff 100644 --- a/api/v1/providers/cohere.json +++ b/api/v1/providers/cohere.json @@ -3,10 +3,10 @@ "docs_url": "https://docs.cohere.com", "models": [ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Aya Expanse 32B", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Aya", "model_id": "aya-expanse-32b", "notes": "Cohere official pricing page states Aya Expanse 8B and 32B API pricing is $0.50/1M input and $1.50/1M output.", @@ -25,14 +25,14 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Command A+", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Command", "model_id": "command-a-plus", - "notes": "Cohere official pricing page lists Command A+ but the extracted public page did not expose a confirmed token price. Values are null rather than guessed.", - "official_source_url": "https://cohere.com/pricing", + "notes": "Cohere's official Command A+ page states that access is free until applicable rate limits are reached and that production use is available through Model Vault. Another official Cohere page shows per-token pricing for the same model ID, so V1 keeps token pricing null until the official documentation is reconciled.", + "official_source_url": "https://docs.cohere.com/docs/command-a-plus", "pricing": { "batch_input": null, "batch_output": null, @@ -47,10 +47,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Command R+ 08-2024", "effective_from": "2024-08-01", - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Command R", "model_id": "command-r-plus-08-2024", "notes": "Cohere official pricing page lists Command R+ 08-2024 at $2.50/1M input tokens and $10.00/1M output tokens.", diff --git a/api/v1/providers/deepseek.json b/api/v1/providers/deepseek.json index b34bff6..16bd173 100644 --- a/api/v1/providers/deepseek.json +++ b/api/v1/providers/deepseek.json @@ -3,10 +3,10 @@ "docs_url": "https://api-docs.deepseek.com", "models": [ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "DeepSeek V4 Flash", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "DeepSeek V4", "model_id": "deepseek-v4-flash", "notes": "DeepSeek labels input cache miss as input price and cache hit as cached_input.", @@ -25,10 +25,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "DeepSeek V4 Pro", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "DeepSeek V4", "model_id": "deepseek-v4-pro", "notes": "DeepSeek labels input cache miss as input price and cache hit as cached_input.", diff --git a/api/v1/providers/google-gemini.json b/api/v1/providers/google-gemini.json index 8a77e2c..e28af5a 100644 --- a/api/v1/providers/google-gemini.json +++ b/api/v1/providers/google-gemini.json @@ -3,10 +3,10 @@ "docs_url": "https://ai.google.dev/gemini-api/docs", "models": [ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Gemini 2.5 Flash", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Gemini 2.5", "model_id": "gemini-2.5-flash", "notes": "Rates are for text/image/video input; Google lists separate audio rates in the official pricing table.", @@ -25,10 +25,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Gemini 2.5 Pro", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Gemini 2.5", "model_id": "gemini-2.5-pro", "notes": "Rates are for prompts up to 200k tokens. Google lists higher rates above 200k tokens in the official pricing table.", diff --git a/api/v1/providers/mistral-ai.json b/api/v1/providers/mistral-ai.json index de8792f..237f155 100644 --- a/api/v1/providers/mistral-ai.json +++ b/api/v1/providers/mistral-ai.json @@ -3,22 +3,22 @@ "docs_url": "https://docs.mistral.ai", "models": [ { - "accessed_at": "2026-07-05T00:00:00Z", - "display_name": "Mistral Large", + "accessed_at": "2026-07-27T00:00:00Z", + "display_name": "Mistral Large 3", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Mistral Large", "model_id": "mistral-large", - "notes": "Mistral official pricing FAQ states Mistral Large costs $2/M input and $6/M output and batch processing gets a 50% discount.", - "official_source_url": "https://mistral.ai/pricing", + "notes": "mistral-large-latest currently maps to Mistral Large 3. Standard price is $0.50/M input and $1.50/M output. Cached input receives a 90% discount. Batch processing receives a 50% discount.", + "official_source_url": "https://mistral.ai/pricing/api/", "pricing": { - "batch_input": 1.0, - "batch_output": 3.0, + "batch_input": 0.25, + "batch_output": 0.75, "cache_write": null, - "cached_input": null, + "cached_input": 0.05, "currency": "USD", - "input": 2.0, - "output": 6.0, + "input": 0.5, + "output": 1.5, "unit": "1M tokens" }, "provider_id": "mistral-ai", diff --git a/api/v1/providers/openai.json b/api/v1/providers/openai.json index cd5be73..e9283aa 100644 --- a/api/v1/providers/openai.json +++ b/api/v1/providers/openai.json @@ -3,10 +3,10 @@ "docs_url": "https://platform.openai.com/docs", "models": [ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-4.1", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-4.1", "model_id": "gpt-4.1", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", @@ -25,10 +25,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5", "model_id": "gpt-5", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", @@ -47,10 +47,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5.4 mini", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5.4", "model_id": "gpt-5.4-mini", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", @@ -69,10 +69,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5.5", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5.5", "model_id": "gpt-5.5", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens. Batch prices are 50% of the listed standard price where shown.", @@ -91,10 +91,10 @@ "status": "active" }, { - "accessed_at": "2026-07-10T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5.6 Luna", "effective_from": null, - "last_verified_at": "2026-07-10T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5.6", "model_id": "gpt-5.6-luna", "notes": "OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices.", @@ -113,10 +113,10 @@ "status": "active" }, { - "accessed_at": "2026-07-10T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5.6 Sol", "effective_from": null, - "last_verified_at": "2026-07-10T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5.6", "model_id": "gpt-5.6-sol", "notes": "OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices.", @@ -135,10 +135,10 @@ "status": "active" }, { - "accessed_at": "2026-07-10T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5.6 Terra", "effective_from": null, - "last_verified_at": "2026-07-10T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5.6", "model_id": "gpt-5.6-terra", "notes": "OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices.", @@ -157,10 +157,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "o3", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "o-series", "model_id": "o3", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", diff --git a/api/v1/providers/xai.json b/api/v1/providers/xai.json index aa642d2..7780370 100644 --- a/api/v1/providers/xai.json +++ b/api/v1/providers/xai.json @@ -3,19 +3,19 @@ "docs_url": "https://docs.x.ai", "models": [ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Grok 4.3", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Grok 4", "model_id": "grok-4.3", - "notes": "xAI official model page lists Grok 4.3 input and output rates per 1M tokens. Cached and batch rates were not shown in the extracted official model card.", - "official_source_url": "https://docs.x.ai/developers/models", + "notes": "V1 stores short-context pricing. The long-context threshold is 200K tokens. Grok 4.3 batch pricing receives a 20% discount. Long-context prices are not stored in the current short-context fields.", + "official_source_url": "https://docs.x.ai/developers/pricing", "pricing": { - "batch_input": null, - "batch_output": null, + "batch_input": 1.0, + "batch_output": 2.0, "cache_write": null, - "cached_input": null, + "cached_input": 0.2, "currency": "USD", "input": 1.25, "output": 2.5, diff --git a/data/canonical/models.json b/data/canonical/models.json index c6bc40a..e653c72 100644 --- a/data/canonical/models.json +++ b/data/canonical/models.json @@ -7,8 +7,8 @@ "status": "active", "pricing": {"currency": "USD", "unit": "1M tokens", "input": 5.0, "output": 30.0, "cached_input": 0.5, "cache_write": 6.25, "batch_input": 2.5, "batch_output": 15.0}, "official_source_url": "https://developers.openai.com/api/docs/pricing", - "accessed_at": "2026-07-10T00:00:00Z", - "last_verified_at": "2026-07-10T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "effective_from": null, "notes": "OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices." }, @@ -20,8 +20,8 @@ "status": "active", "pricing": {"currency": "USD", "unit": "1M tokens", "input": 2.5, "output": 15.0, "cached_input": 0.25, "cache_write": 3.125, "batch_input": 1.25, "batch_output": 7.5}, "official_source_url": "https://developers.openai.com/api/docs/pricing", - "accessed_at": "2026-07-10T00:00:00Z", - "last_verified_at": "2026-07-10T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "effective_from": null, "notes": "OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices." }, @@ -33,8 +33,8 @@ "status": "active", "pricing": {"currency": "USD", "unit": "1M tokens", "input": 1.0, "output": 6.0, "cached_input": 0.1, "cache_write": 1.25, "batch_input": 0.5, "batch_output": 3.0}, "official_source_url": "https://developers.openai.com/api/docs/pricing", - "accessed_at": "2026-07-10T00:00:00Z", - "last_verified_at": "2026-07-10T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "effective_from": null, "notes": "OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices." }, @@ -46,8 +46,8 @@ "status": "active", "pricing": {"currency": "USD", "unit": "1M tokens", "input": 5.0, "output": 30.0, "cached_input": 0.5, "cache_write": null, "batch_input": 2.5, "batch_output": 15.0}, "official_source_url": "https://platform.openai.com/docs/pricing", - "accessed_at": "2026-07-05T00:00:00Z", - "last_verified_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "effective_from": null, "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens. Batch prices are 50% of the listed standard price where shown." }, @@ -59,8 +59,8 @@ "status": "active", "pricing": {"currency": "USD", "unit": "1M tokens", "input": 0.75, "output": 4.5, "cached_input": 0.075, "cache_write": null, "batch_input": 0.375, "batch_output": 2.25}, "official_source_url": "https://platform.openai.com/docs/pricing", - "accessed_at": "2026-07-05T00:00:00Z", - "last_verified_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "effective_from": null, "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens." }, @@ -72,8 +72,8 @@ "status": "active", "pricing": {"currency": "USD", "unit": "1M tokens", "input": 1.25, "output": 10.0, "cached_input": 0.125, "cache_write": null, "batch_input": 0.625, "batch_output": 5.0}, "official_source_url": "https://platform.openai.com/docs/pricing", - "accessed_at": "2026-07-05T00:00:00Z", - "last_verified_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "effective_from": null, "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens." }, @@ -85,8 +85,8 @@ "status": "active", "pricing": {"currency": "USD", "unit": "1M tokens", "input": 2.0, "output": 8.0, "cached_input": 0.5, "cache_write": null, "batch_input": 1.0, "batch_output": 4.0}, "official_source_url": "https://platform.openai.com/docs/pricing", - "accessed_at": "2026-07-05T00:00:00Z", - "last_verified_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "effective_from": null, "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens." }, @@ -98,8 +98,8 @@ "status": "active", "pricing": {"currency": "USD", "unit": "1M tokens", "input": 2.0, "output": 8.0, "cached_input": 0.5, "cache_write": null, "batch_input": 1.0, "batch_output": 4.0}, "official_source_url": "https://platform.openai.com/docs/pricing", - "accessed_at": "2026-07-05T00:00:00Z", - "last_verified_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "effective_from": null, "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens." }, @@ -111,8 +111,8 @@ "status": "active", "pricing": {"currency": "USD", "unit": "1M tokens", "input": 5.0, "output": 25.0, "cached_input": 0.5, "cache_write": 6.25, "batch_input": 2.5, "batch_output": 12.5}, "official_source_url": "https://docs.anthropic.com/en/docs/about-claude/pricing", - "accessed_at": "2026-07-05T00:00:00Z", - "last_verified_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "effective_from": null, "notes": "Anthropic table columns are input, 5-minute cache write, 1-hour cache write, cache read, and output. cache_write stores the 5-minute write rate." }, @@ -123,11 +123,11 @@ "model_family": "Claude Sonnet", "status": "active", "pricing": {"currency": "USD", "unit": "1M tokens", "input": 2.0, "output": 10.0, "cached_input": 0.2, "cache_write": 2.5, "batch_input": 1.0, "batch_output": 5.0}, - "official_source_url": "https://docs.anthropic.com/en/docs/about-claude/pricing#claude-sonnet-5-introductory-pricing", - "accessed_at": "2026-07-05T00:00:00Z", - "last_verified_at": "2026-07-05T00:00:00Z", - "effective_from": "2026-07-05", - "notes": "Introductory pricing is listed by Anthropic as effective through 2026-08-31; standard pricing starts 2026-09-01." + "official_source_url": "https://www.anthropic.com/news/claude-sonnet-5", + "accessed_at": "2026-07-27T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", + "effective_from": "2026-06-30", + "notes": "Introductory pricing applies through 2026-08-31 and standard pricing starts 2026-09-01. V1 must not switch early to the standard $3/M input and $15/M output rates." }, { "provider_id": "anthropic", @@ -137,8 +137,8 @@ "status": "active", "pricing": {"currency": "USD", "unit": "1M tokens", "input": 3.0, "output": 15.0, "cached_input": 0.3, "cache_write": 3.75, "batch_input": 1.5, "batch_output": 7.5}, "official_source_url": "https://docs.anthropic.com/en/docs/about-claude/pricing", - "accessed_at": "2026-07-05T00:00:00Z", - "last_verified_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "effective_from": null, "notes": "cache_write stores the 5-minute write rate." }, @@ -150,8 +150,8 @@ "status": "active", "pricing": {"currency": "USD", "unit": "1M tokens", "input": 1.0, "output": 5.0, "cached_input": 0.1, "cache_write": 1.25, "batch_input": 0.5, "batch_output": 2.5}, "official_source_url": "https://docs.anthropic.com/en/docs/about-claude/pricing", - "accessed_at": "2026-07-05T00:00:00Z", - "last_verified_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "effective_from": null, "notes": "cache_write stores the 5-minute write rate." }, @@ -163,8 +163,8 @@ "status": "active", "pricing": {"currency": "USD", "unit": "1M tokens", "input": 1.25, "output": 10.0, "cached_input": 0.125, "cache_write": null, "batch_input": 0.625, "batch_output": 5.0}, "official_source_url": "https://ai.google.dev/gemini-api/docs/pricing", - "accessed_at": "2026-07-05T00:00:00Z", - "last_verified_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "effective_from": null, "notes": "Rates are for prompts up to 200k tokens. Google lists higher rates above 200k tokens in the official pricing table." }, @@ -176,8 +176,8 @@ "status": "active", "pricing": {"currency": "USD", "unit": "1M tokens", "input": 0.3, "output": 2.5, "cached_input": 0.03, "cache_write": null, "batch_input": 0.15, "batch_output": 1.25}, "official_source_url": "https://ai.google.dev/gemini-api/docs/pricing", - "accessed_at": "2026-07-05T00:00:00Z", - "last_verified_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "effective_from": null, "notes": "Rates are for text/image/video input; Google lists separate audio rates in the official pricing table." }, @@ -187,12 +187,12 @@ "display_name": "Grok 4.3", "model_family": "Grok 4", "status": "active", - "pricing": {"currency": "USD", "unit": "1M tokens", "input": 1.25, "output": 2.5, "cached_input": null, "cache_write": null, "batch_input": null, "batch_output": null}, - "official_source_url": "https://docs.x.ai/developers/models", - "accessed_at": "2026-07-05T00:00:00Z", - "last_verified_at": "2026-07-05T00:00:00Z", + "pricing": {"currency": "USD", "unit": "1M tokens", "input": 1.25, "output": 2.5, "cached_input": 0.2, "cache_write": null, "batch_input": 1.0, "batch_output": 2.0}, + "official_source_url": "https://docs.x.ai/developers/pricing", + "accessed_at": "2026-07-27T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "effective_from": null, - "notes": "xAI official model page lists Grok 4.3 input and output rates per 1M tokens. Cached and batch rates were not shown in the extracted official model card." + "notes": "V1 stores short-context pricing. The long-context threshold is 200K tokens. Grok 4.3 batch pricing receives a 20% discount. Long-context prices are not stored in the current short-context fields." }, { "provider_id": "deepseek", @@ -202,8 +202,8 @@ "status": "active", "pricing": {"currency": "USD", "unit": "1M tokens", "input": 0.14, "output": 0.28, "cached_input": 0.0028, "cache_write": null, "batch_input": null, "batch_output": null}, "official_source_url": "https://api-docs.deepseek.com/quick_start/pricing", - "accessed_at": "2026-07-05T00:00:00Z", - "last_verified_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "effective_from": null, "notes": "DeepSeek labels input cache miss as input price and cache hit as cached_input." }, @@ -215,23 +215,23 @@ "status": "active", "pricing": {"currency": "USD", "unit": "1M tokens", "input": 0.435, "output": 0.87, "cached_input": 0.003625, "cache_write": null, "batch_input": null, "batch_output": null}, "official_source_url": "https://api-docs.deepseek.com/quick_start/pricing", - "accessed_at": "2026-07-05T00:00:00Z", - "last_verified_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "effective_from": null, "notes": "DeepSeek labels input cache miss as input price and cache hit as cached_input." }, { "provider_id": "mistral-ai", "model_id": "mistral-large", - "display_name": "Mistral Large", + "display_name": "Mistral Large 3", "model_family": "Mistral Large", "status": "active", - "pricing": {"currency": "USD", "unit": "1M tokens", "input": 2.0, "output": 6.0, "cached_input": null, "cache_write": null, "batch_input": 1.0, "batch_output": 3.0}, - "official_source_url": "https://mistral.ai/pricing", - "accessed_at": "2026-07-05T00:00:00Z", - "last_verified_at": "2026-07-05T00:00:00Z", + "pricing": {"currency": "USD", "unit": "1M tokens", "input": 0.5, "output": 1.5, "cached_input": 0.05, "cache_write": null, "batch_input": 0.25, "batch_output": 0.75}, + "official_source_url": "https://mistral.ai/pricing/api/", + "accessed_at": "2026-07-27T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "effective_from": null, - "notes": "Mistral official pricing FAQ states Mistral Large costs $2/M input and $6/M output and batch processing gets a 50% discount." + "notes": "mistral-large-latest currently maps to Mistral Large 3. Standard price is $0.50/M input and $1.50/M output. Cached input receives a 90% discount. Batch processing receives a 50% discount." }, { "provider_id": "cohere", @@ -240,11 +240,11 @@ "model_family": "Command", "status": "active", "pricing": {"currency": "USD", "unit": "1M tokens", "input": null, "output": null, "cached_input": null, "cache_write": null, "batch_input": null, "batch_output": null}, - "official_source_url": "https://cohere.com/pricing", - "accessed_at": "2026-07-05T00:00:00Z", - "last_verified_at": "2026-07-05T00:00:00Z", + "official_source_url": "https://docs.cohere.com/docs/command-a-plus", + "accessed_at": "2026-07-27T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "effective_from": null, - "notes": "Cohere official pricing page lists Command A+ but the extracted public page did not expose a confirmed token price. Values are null rather than guessed." + "notes": "Cohere's official Command A+ page states that access is free until applicable rate limits are reached and that production use is available through Model Vault. Another official Cohere page shows per-token pricing for the same model ID, so V1 keeps token pricing null until the official documentation is reconciled." }, { "provider_id": "cohere", @@ -254,8 +254,8 @@ "status": "active", "pricing": {"currency": "USD", "unit": "1M tokens", "input": 2.5, "output": 10.0, "cached_input": null, "cache_write": null, "batch_input": null, "batch_output": null}, "official_source_url": "https://cohere.com/pricing", - "accessed_at": "2026-07-05T00:00:00Z", - "last_verified_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "effective_from": "2024-08-01", "notes": "Cohere official pricing page lists Command R+ 08-2024 at $2.50/1M input tokens and $10.00/1M output tokens." }, @@ -267,8 +267,8 @@ "status": "active", "pricing": {"currency": "USD", "unit": "1M tokens", "input": 0.5, "output": 1.5, "cached_input": null, "cache_write": null, "batch_input": null, "batch_output": null}, "official_source_url": "https://cohere.com/pricing", - "accessed_at": "2026-07-05T00:00:00Z", - "last_verified_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "effective_from": null, "notes": "Cohere official pricing page states Aya Expanse 8B and 32B API pricing is $0.50/1M input and $1.50/1M output." } diff --git a/data/history/anthropic/claude-haiku-4.5.jsonl b/data/history/anthropic/claude-haiku-4.5.jsonl index 2d5d0f9..1bf1637 100644 --- a/data/history/anthropic/claude-haiku-4.5.jsonl +++ b/data/history/anthropic/claude-haiku-4.5.jsonl @@ -1 +1,2 @@ {"effective_from": null, "last_verified_at": "2026-07-05T00:00:00Z", "model_id": "claude-haiku-4.5", "notes": "cache_write stores the 5-minute write rate.", "official_source_url": "https://docs.anthropic.com/en/docs/about-claude/pricing", "pricing": {"batch_input": 0.5, "batch_output": 2.5, "cache_write": 1.25, "cached_input": 0.1, "currency": "USD", "input": 1.0, "output": 5.0, "unit": "1M tokens"}, "provider_id": "anthropic", "recorded_at": "2026-07-05T00:00:00Z"} +{"effective_from": null, "last_verified_at": "2026-07-27T00:00:00Z", "model_id": "claude-haiku-4.5", "notes": "cache_write stores the 5-minute write rate.", "official_source_url": "https://docs.anthropic.com/en/docs/about-claude/pricing", "pricing": {"batch_input": 0.5, "batch_output": 2.5, "cache_write": 1.25, "cached_input": 0.1, "currency": "USD", "input": 1.0, "output": 5.0, "unit": "1M tokens"}, "provider_id": "anthropic", "recorded_at": "2026-07-27T00:00:00Z"} diff --git a/data/history/anthropic/claude-opus-4.8.jsonl b/data/history/anthropic/claude-opus-4.8.jsonl index 474c28a..72cd343 100644 --- a/data/history/anthropic/claude-opus-4.8.jsonl +++ b/data/history/anthropic/claude-opus-4.8.jsonl @@ -1 +1,2 @@ {"effective_from": null, "last_verified_at": "2026-07-05T00:00:00Z", "model_id": "claude-opus-4.8", "notes": "Anthropic table columns are input, 5-minute cache write, 1-hour cache write, cache read, and output. cache_write stores the 5-minute write rate.", "official_source_url": "https://docs.anthropic.com/en/docs/about-claude/pricing", "pricing": {"batch_input": 2.5, "batch_output": 12.5, "cache_write": 6.25, "cached_input": 0.5, "currency": "USD", "input": 5.0, "output": 25.0, "unit": "1M tokens"}, "provider_id": "anthropic", "recorded_at": "2026-07-05T00:00:00Z"} +{"effective_from": null, "last_verified_at": "2026-07-27T00:00:00Z", "model_id": "claude-opus-4.8", "notes": "Anthropic table columns are input, 5-minute cache write, 1-hour cache write, cache read, and output. cache_write stores the 5-minute write rate.", "official_source_url": "https://docs.anthropic.com/en/docs/about-claude/pricing", "pricing": {"batch_input": 2.5, "batch_output": 12.5, "cache_write": 6.25, "cached_input": 0.5, "currency": "USD", "input": 5.0, "output": 25.0, "unit": "1M tokens"}, "provider_id": "anthropic", "recorded_at": "2026-07-27T00:00:00Z"} diff --git a/data/history/anthropic/claude-sonnet-4.6.jsonl b/data/history/anthropic/claude-sonnet-4.6.jsonl index 2f87181..3031358 100644 --- a/data/history/anthropic/claude-sonnet-4.6.jsonl +++ b/data/history/anthropic/claude-sonnet-4.6.jsonl @@ -1 +1,2 @@ {"effective_from": null, "last_verified_at": "2026-07-05T00:00:00Z", "model_id": "claude-sonnet-4.6", "notes": "cache_write stores the 5-minute write rate.", "official_source_url": "https://docs.anthropic.com/en/docs/about-claude/pricing", "pricing": {"batch_input": 1.5, "batch_output": 7.5, "cache_write": 3.75, "cached_input": 0.3, "currency": "USD", "input": 3.0, "output": 15.0, "unit": "1M tokens"}, "provider_id": "anthropic", "recorded_at": "2026-07-05T00:00:00Z"} +{"effective_from": null, "last_verified_at": "2026-07-27T00:00:00Z", "model_id": "claude-sonnet-4.6", "notes": "cache_write stores the 5-minute write rate.", "official_source_url": "https://docs.anthropic.com/en/docs/about-claude/pricing", "pricing": {"batch_input": 1.5, "batch_output": 7.5, "cache_write": 3.75, "cached_input": 0.3, "currency": "USD", "input": 3.0, "output": 15.0, "unit": "1M tokens"}, "provider_id": "anthropic", "recorded_at": "2026-07-27T00:00:00Z"} diff --git a/data/history/anthropic/claude-sonnet-5-intro.jsonl b/data/history/anthropic/claude-sonnet-5-intro.jsonl index c9423db..e5f9a5c 100644 --- a/data/history/anthropic/claude-sonnet-5-intro.jsonl +++ b/data/history/anthropic/claude-sonnet-5-intro.jsonl @@ -1 +1,2 @@ {"effective_from": "2026-07-05", "last_verified_at": "2026-07-05T00:00:00Z", "model_id": "claude-sonnet-5-intro", "notes": "Introductory pricing is listed by Anthropic as effective through 2026-08-31; standard pricing starts 2026-09-01.", "official_source_url": "https://docs.anthropic.com/en/docs/about-claude/pricing#claude-sonnet-5-introductory-pricing", "pricing": {"batch_input": 1.0, "batch_output": 5.0, "cache_write": 2.5, "cached_input": 0.2, "currency": "USD", "input": 2.0, "output": 10.0, "unit": "1M tokens"}, "provider_id": "anthropic", "recorded_at": "2026-07-05T00:00:00Z"} +{"effective_from": "2026-06-30", "last_verified_at": "2026-07-27T00:00:00Z", "model_id": "claude-sonnet-5-intro", "notes": "Introductory pricing applies through 2026-08-31 and standard pricing starts 2026-09-01. V1 must not switch early to the standard $3/M input and $15/M output rates.", "official_source_url": "https://www.anthropic.com/news/claude-sonnet-5", "pricing": {"batch_input": 1.0, "batch_output": 5.0, "cache_write": 2.5, "cached_input": 0.2, "currency": "USD", "input": 2.0, "output": 10.0, "unit": "1M tokens"}, "provider_id": "anthropic", "recorded_at": "2026-07-27T00:00:00Z"} diff --git a/data/history/cohere/aya-expanse-32b.jsonl b/data/history/cohere/aya-expanse-32b.jsonl index d20826d..37a9961 100644 --- a/data/history/cohere/aya-expanse-32b.jsonl +++ b/data/history/cohere/aya-expanse-32b.jsonl @@ -1 +1,2 @@ {"effective_from": null, "last_verified_at": "2026-07-05T00:00:00Z", "model_id": "aya-expanse-32b", "notes": "Cohere official pricing page states Aya Expanse 8B and 32B API pricing is $0.50/1M input and $1.50/1M output.", "official_source_url": "https://cohere.com/pricing", "pricing": {"batch_input": null, "batch_output": null, "cache_write": null, "cached_input": null, "currency": "USD", "input": 0.5, "output": 1.5, "unit": "1M tokens"}, "provider_id": "cohere", "recorded_at": "2026-07-05T00:00:00Z"} +{"effective_from": null, "last_verified_at": "2026-07-27T00:00:00Z", "model_id": "aya-expanse-32b", "notes": "Cohere official pricing page states Aya Expanse 8B and 32B API pricing is $0.50/1M input and $1.50/1M output.", "official_source_url": "https://cohere.com/pricing", "pricing": {"batch_input": null, "batch_output": null, "cache_write": null, "cached_input": null, "currency": "USD", "input": 0.5, "output": 1.5, "unit": "1M tokens"}, "provider_id": "cohere", "recorded_at": "2026-07-27T00:00:00Z"} diff --git a/data/history/cohere/command-a-plus.jsonl b/data/history/cohere/command-a-plus.jsonl index d23689b..b08d9ea 100644 --- a/data/history/cohere/command-a-plus.jsonl +++ b/data/history/cohere/command-a-plus.jsonl @@ -1 +1,2 @@ {"effective_from": null, "last_verified_at": "2026-07-05T00:00:00Z", "model_id": "command-a-plus", "notes": "Cohere official pricing page lists Command A+ but the extracted public page did not expose a confirmed token price. Values are null rather than guessed.", "official_source_url": "https://cohere.com/pricing", "pricing": {"batch_input": null, "batch_output": null, "cache_write": null, "cached_input": null, "currency": "USD", "input": null, "output": null, "unit": "1M tokens"}, "provider_id": "cohere", "recorded_at": "2026-07-05T00:00:00Z"} +{"effective_from": null, "last_verified_at": "2026-07-27T00:00:00Z", "model_id": "command-a-plus", "notes": "Cohere's official Command A+ page states that access is free until applicable rate limits are reached and that production use is available through Model Vault. Another official Cohere page shows per-token pricing for the same model ID, so V1 keeps token pricing null until the official documentation is reconciled.", "official_source_url": "https://docs.cohere.com/docs/command-a-plus", "pricing": {"batch_input": null, "batch_output": null, "cache_write": null, "cached_input": null, "currency": "USD", "input": null, "output": null, "unit": "1M tokens"}, "provider_id": "cohere", "recorded_at": "2026-07-27T00:00:00Z"} diff --git a/data/history/cohere/command-r-plus-08-2024.jsonl b/data/history/cohere/command-r-plus-08-2024.jsonl index c8cd519..4b5cfb7 100644 --- a/data/history/cohere/command-r-plus-08-2024.jsonl +++ b/data/history/cohere/command-r-plus-08-2024.jsonl @@ -1 +1,2 @@ {"effective_from": "2024-08-01", "last_verified_at": "2026-07-05T00:00:00Z", "model_id": "command-r-plus-08-2024", "notes": "Cohere official pricing page lists Command R+ 08-2024 at $2.50/1M input tokens and $10.00/1M output tokens.", "official_source_url": "https://cohere.com/pricing", "pricing": {"batch_input": null, "batch_output": null, "cache_write": null, "cached_input": null, "currency": "USD", "input": 2.5, "output": 10.0, "unit": "1M tokens"}, "provider_id": "cohere", "recorded_at": "2026-07-05T00:00:00Z"} +{"effective_from": "2024-08-01", "last_verified_at": "2026-07-27T00:00:00Z", "model_id": "command-r-plus-08-2024", "notes": "Cohere official pricing page lists Command R+ 08-2024 at $2.50/1M input tokens and $10.00/1M output tokens.", "official_source_url": "https://cohere.com/pricing", "pricing": {"batch_input": null, "batch_output": null, "cache_write": null, "cached_input": null, "currency": "USD", "input": 2.5, "output": 10.0, "unit": "1M tokens"}, "provider_id": "cohere", "recorded_at": "2026-07-27T00:00:00Z"} diff --git a/data/history/deepseek/deepseek-v4-flash.jsonl b/data/history/deepseek/deepseek-v4-flash.jsonl index 4129c56..0c33ed3 100644 --- a/data/history/deepseek/deepseek-v4-flash.jsonl +++ b/data/history/deepseek/deepseek-v4-flash.jsonl @@ -1 +1,2 @@ {"effective_from": null, "last_verified_at": "2026-07-05T00:00:00Z", "model_id": "deepseek-v4-flash", "notes": "DeepSeek labels input cache miss as input price and cache hit as cached_input.", "official_source_url": "https://api-docs.deepseek.com/quick_start/pricing", "pricing": {"batch_input": null, "batch_output": null, "cache_write": null, "cached_input": 0.0028, "currency": "USD", "input": 0.14, "output": 0.28, "unit": "1M tokens"}, "provider_id": "deepseek", "recorded_at": "2026-07-05T00:00:00Z"} +{"effective_from": null, "last_verified_at": "2026-07-27T00:00:00Z", "model_id": "deepseek-v4-flash", "notes": "DeepSeek labels input cache miss as input price and cache hit as cached_input.", "official_source_url": "https://api-docs.deepseek.com/quick_start/pricing", "pricing": {"batch_input": null, "batch_output": null, "cache_write": null, "cached_input": 0.0028, "currency": "USD", "input": 0.14, "output": 0.28, "unit": "1M tokens"}, "provider_id": "deepseek", "recorded_at": "2026-07-27T00:00:00Z"} diff --git a/data/history/deepseek/deepseek-v4-pro.jsonl b/data/history/deepseek/deepseek-v4-pro.jsonl index 7223967..8af6738 100644 --- a/data/history/deepseek/deepseek-v4-pro.jsonl +++ b/data/history/deepseek/deepseek-v4-pro.jsonl @@ -1 +1,2 @@ {"effective_from": null, "last_verified_at": "2026-07-05T00:00:00Z", "model_id": "deepseek-v4-pro", "notes": "DeepSeek labels input cache miss as input price and cache hit as cached_input.", "official_source_url": "https://api-docs.deepseek.com/quick_start/pricing", "pricing": {"batch_input": null, "batch_output": null, "cache_write": null, "cached_input": 0.003625, "currency": "USD", "input": 0.435, "output": 0.87, "unit": "1M tokens"}, "provider_id": "deepseek", "recorded_at": "2026-07-05T00:00:00Z"} +{"effective_from": null, "last_verified_at": "2026-07-27T00:00:00Z", "model_id": "deepseek-v4-pro", "notes": "DeepSeek labels input cache miss as input price and cache hit as cached_input.", "official_source_url": "https://api-docs.deepseek.com/quick_start/pricing", "pricing": {"batch_input": null, "batch_output": null, "cache_write": null, "cached_input": 0.003625, "currency": "USD", "input": 0.435, "output": 0.87, "unit": "1M tokens"}, "provider_id": "deepseek", "recorded_at": "2026-07-27T00:00:00Z"} diff --git a/data/history/google-gemini/gemini-2.5-flash.jsonl b/data/history/google-gemini/gemini-2.5-flash.jsonl index 183dbb2..f346aef 100644 --- a/data/history/google-gemini/gemini-2.5-flash.jsonl +++ b/data/history/google-gemini/gemini-2.5-flash.jsonl @@ -1 +1,2 @@ {"effective_from": null, "last_verified_at": "2026-07-05T00:00:00Z", "model_id": "gemini-2.5-flash", "notes": "Rates are for text/image/video input; Google lists separate audio rates in the official pricing table.", "official_source_url": "https://ai.google.dev/gemini-api/docs/pricing", "pricing": {"batch_input": 0.15, "batch_output": 1.25, "cache_write": null, "cached_input": 0.03, "currency": "USD", "input": 0.3, "output": 2.5, "unit": "1M tokens"}, "provider_id": "google-gemini", "recorded_at": "2026-07-05T00:00:00Z"} +{"effective_from": null, "last_verified_at": "2026-07-27T00:00:00Z", "model_id": "gemini-2.5-flash", "notes": "Rates are for text/image/video input; Google lists separate audio rates in the official pricing table.", "official_source_url": "https://ai.google.dev/gemini-api/docs/pricing", "pricing": {"batch_input": 0.15, "batch_output": 1.25, "cache_write": null, "cached_input": 0.03, "currency": "USD", "input": 0.3, "output": 2.5, "unit": "1M tokens"}, "provider_id": "google-gemini", "recorded_at": "2026-07-27T00:00:00Z"} diff --git a/data/history/google-gemini/gemini-2.5-pro.jsonl b/data/history/google-gemini/gemini-2.5-pro.jsonl index 9b6aa70..d70cf17 100644 --- a/data/history/google-gemini/gemini-2.5-pro.jsonl +++ b/data/history/google-gemini/gemini-2.5-pro.jsonl @@ -1 +1,2 @@ {"effective_from": null, "last_verified_at": "2026-07-05T00:00:00Z", "model_id": "gemini-2.5-pro", "notes": "Rates are for prompts up to 200k tokens. Google lists higher rates above 200k tokens in the official pricing table.", "official_source_url": "https://ai.google.dev/gemini-api/docs/pricing", "pricing": {"batch_input": 0.625, "batch_output": 5.0, "cache_write": null, "cached_input": 0.125, "currency": "USD", "input": 1.25, "output": 10.0, "unit": "1M tokens"}, "provider_id": "google-gemini", "recorded_at": "2026-07-05T00:00:00Z"} +{"effective_from": null, "last_verified_at": "2026-07-27T00:00:00Z", "model_id": "gemini-2.5-pro", "notes": "Rates are for prompts up to 200k tokens. Google lists higher rates above 200k tokens in the official pricing table.", "official_source_url": "https://ai.google.dev/gemini-api/docs/pricing", "pricing": {"batch_input": 0.625, "batch_output": 5.0, "cache_write": null, "cached_input": 0.125, "currency": "USD", "input": 1.25, "output": 10.0, "unit": "1M tokens"}, "provider_id": "google-gemini", "recorded_at": "2026-07-27T00:00:00Z"} diff --git a/data/history/mistral-ai/mistral-large.jsonl b/data/history/mistral-ai/mistral-large.jsonl index 1377e31..aa842ae 100644 --- a/data/history/mistral-ai/mistral-large.jsonl +++ b/data/history/mistral-ai/mistral-large.jsonl @@ -1 +1,2 @@ {"effective_from": null, "last_verified_at": "2026-07-05T00:00:00Z", "model_id": "mistral-large", "notes": "Mistral official pricing FAQ states Mistral Large costs $2/M input and $6/M output and batch processing gets a 50% discount.", "official_source_url": "https://mistral.ai/pricing", "pricing": {"batch_input": 1.0, "batch_output": 3.0, "cache_write": null, "cached_input": null, "currency": "USD", "input": 2.0, "output": 6.0, "unit": "1M tokens"}, "provider_id": "mistral-ai", "recorded_at": "2026-07-05T00:00:00Z"} +{"effective_from": null, "last_verified_at": "2026-07-27T00:00:00Z", "model_id": "mistral-large", "notes": "mistral-large-latest currently maps to Mistral Large 3. Standard price is $0.50/M input and $1.50/M output. Cached input receives a 90% discount. Batch processing receives a 50% discount.", "official_source_url": "https://mistral.ai/pricing/api/", "pricing": {"batch_input": 0.25, "batch_output": 0.75, "cache_write": null, "cached_input": 0.05, "currency": "USD", "input": 0.5, "output": 1.5, "unit": "1M tokens"}, "provider_id": "mistral-ai", "recorded_at": "2026-07-27T00:00:00Z"} diff --git a/data/history/openai/gpt-4.1.jsonl b/data/history/openai/gpt-4.1.jsonl index 04118a8..326237c 100644 --- a/data/history/openai/gpt-4.1.jsonl +++ b/data/history/openai/gpt-4.1.jsonl @@ -1 +1,2 @@ {"effective_from": null, "last_verified_at": "2026-07-05T00:00:00Z", "model_id": "gpt-4.1", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", "official_source_url": "https://platform.openai.com/docs/pricing", "pricing": {"batch_input": 1.0, "batch_output": 4.0, "cache_write": null, "cached_input": 0.5, "currency": "USD", "input": 2.0, "output": 8.0, "unit": "1M tokens"}, "provider_id": "openai", "recorded_at": "2026-07-05T00:00:00Z"} +{"effective_from": null, "last_verified_at": "2026-07-27T00:00:00Z", "model_id": "gpt-4.1", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", "official_source_url": "https://platform.openai.com/docs/pricing", "pricing": {"batch_input": 1.0, "batch_output": 4.0, "cache_write": null, "cached_input": 0.5, "currency": "USD", "input": 2.0, "output": 8.0, "unit": "1M tokens"}, "provider_id": "openai", "recorded_at": "2026-07-27T00:00:00Z"} diff --git a/data/history/openai/gpt-5.4-mini.jsonl b/data/history/openai/gpt-5.4-mini.jsonl index de7d7d7..66d95d2 100644 --- a/data/history/openai/gpt-5.4-mini.jsonl +++ b/data/history/openai/gpt-5.4-mini.jsonl @@ -1 +1,2 @@ {"effective_from": null, "last_verified_at": "2026-07-05T00:00:00Z", "model_id": "gpt-5.4-mini", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", "official_source_url": "https://platform.openai.com/docs/pricing", "pricing": {"batch_input": 0.375, "batch_output": 2.25, "cache_write": null, "cached_input": 0.075, "currency": "USD", "input": 0.75, "output": 4.5, "unit": "1M tokens"}, "provider_id": "openai", "recorded_at": "2026-07-05T00:00:00Z"} +{"effective_from": null, "last_verified_at": "2026-07-27T00:00:00Z", "model_id": "gpt-5.4-mini", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", "official_source_url": "https://platform.openai.com/docs/pricing", "pricing": {"batch_input": 0.375, "batch_output": 2.25, "cache_write": null, "cached_input": 0.075, "currency": "USD", "input": 0.75, "output": 4.5, "unit": "1M tokens"}, "provider_id": "openai", "recorded_at": "2026-07-27T00:00:00Z"} diff --git a/data/history/openai/gpt-5.5.jsonl b/data/history/openai/gpt-5.5.jsonl index 599aed4..41df53e 100644 --- a/data/history/openai/gpt-5.5.jsonl +++ b/data/history/openai/gpt-5.5.jsonl @@ -1 +1,2 @@ {"effective_from": null, "last_verified_at": "2026-07-05T00:00:00Z", "model_id": "gpt-5.5", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens. Batch prices are 50% of the listed standard price where shown.", "official_source_url": "https://platform.openai.com/docs/pricing", "pricing": {"batch_input": 2.5, "batch_output": 15.0, "cache_write": null, "cached_input": 0.5, "currency": "USD", "input": 5.0, "output": 30.0, "unit": "1M tokens"}, "provider_id": "openai", "recorded_at": "2026-07-05T00:00:00Z"} +{"effective_from": null, "last_verified_at": "2026-07-27T00:00:00Z", "model_id": "gpt-5.5", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens. Batch prices are 50% of the listed standard price where shown.", "official_source_url": "https://platform.openai.com/docs/pricing", "pricing": {"batch_input": 2.5, "batch_output": 15.0, "cache_write": null, "cached_input": 0.5, "currency": "USD", "input": 5.0, "output": 30.0, "unit": "1M tokens"}, "provider_id": "openai", "recorded_at": "2026-07-27T00:00:00Z"} diff --git a/data/history/openai/gpt-5.6-luna.jsonl b/data/history/openai/gpt-5.6-luna.jsonl index 9ab1ec3..5600655 100644 --- a/data/history/openai/gpt-5.6-luna.jsonl +++ b/data/history/openai/gpt-5.6-luna.jsonl @@ -1 +1,2 @@ {"effective_from": null, "last_verified_at": "2026-07-10T00:00:00Z", "model_id": "gpt-5.6-luna", "notes": "OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices.", "official_source_url": "https://developers.openai.com/api/docs/pricing", "pricing": {"batch_input": 0.5, "batch_output": 3.0, "cache_write": 1.25, "cached_input": 0.1, "currency": "USD", "input": 1.0, "output": 6.0, "unit": "1M tokens"}, "provider_id": "openai", "recorded_at": "2026-07-09T18:36:17Z"} +{"effective_from": null, "last_verified_at": "2026-07-27T00:00:00Z", "model_id": "gpt-5.6-luna", "notes": "OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices.", "official_source_url": "https://developers.openai.com/api/docs/pricing", "pricing": {"batch_input": 0.5, "batch_output": 3.0, "cache_write": 1.25, "cached_input": 0.1, "currency": "USD", "input": 1.0, "output": 6.0, "unit": "1M tokens"}, "provider_id": "openai", "recorded_at": "2026-07-27T00:00:00Z"} diff --git a/data/history/openai/gpt-5.6-sol.jsonl b/data/history/openai/gpt-5.6-sol.jsonl index 54713e9..571d713 100644 --- a/data/history/openai/gpt-5.6-sol.jsonl +++ b/data/history/openai/gpt-5.6-sol.jsonl @@ -1 +1,2 @@ {"effective_from": null, "last_verified_at": "2026-07-10T00:00:00Z", "model_id": "gpt-5.6-sol", "notes": "OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices.", "official_source_url": "https://developers.openai.com/api/docs/pricing", "pricing": {"batch_input": 2.5, "batch_output": 15.0, "cache_write": 6.25, "cached_input": 0.5, "currency": "USD", "input": 5.0, "output": 30.0, "unit": "1M tokens"}, "provider_id": "openai", "recorded_at": "2026-07-09T18:36:17Z"} +{"effective_from": null, "last_verified_at": "2026-07-27T00:00:00Z", "model_id": "gpt-5.6-sol", "notes": "OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices.", "official_source_url": "https://developers.openai.com/api/docs/pricing", "pricing": {"batch_input": 2.5, "batch_output": 15.0, "cache_write": 6.25, "cached_input": 0.5, "currency": "USD", "input": 5.0, "output": 30.0, "unit": "1M tokens"}, "provider_id": "openai", "recorded_at": "2026-07-27T00:00:00Z"} diff --git a/data/history/openai/gpt-5.6-terra.jsonl b/data/history/openai/gpt-5.6-terra.jsonl index 5cbd273..99bad16 100644 --- a/data/history/openai/gpt-5.6-terra.jsonl +++ b/data/history/openai/gpt-5.6-terra.jsonl @@ -1 +1,2 @@ {"effective_from": null, "last_verified_at": "2026-07-10T00:00:00Z", "model_id": "gpt-5.6-terra", "notes": "OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices.", "official_source_url": "https://developers.openai.com/api/docs/pricing", "pricing": {"batch_input": 1.25, "batch_output": 7.5, "cache_write": 3.125, "cached_input": 0.25, "currency": "USD", "input": 2.5, "output": 15.0, "unit": "1M tokens"}, "provider_id": "openai", "recorded_at": "2026-07-09T18:36:17Z"} +{"effective_from": null, "last_verified_at": "2026-07-27T00:00:00Z", "model_id": "gpt-5.6-terra", "notes": "OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices.", "official_source_url": "https://developers.openai.com/api/docs/pricing", "pricing": {"batch_input": 1.25, "batch_output": 7.5, "cache_write": 3.125, "cached_input": 0.25, "currency": "USD", "input": 2.5, "output": 15.0, "unit": "1M tokens"}, "provider_id": "openai", "recorded_at": "2026-07-27T00:00:00Z"} diff --git a/data/history/openai/gpt-5.jsonl b/data/history/openai/gpt-5.jsonl index ba82f63..2322150 100644 --- a/data/history/openai/gpt-5.jsonl +++ b/data/history/openai/gpt-5.jsonl @@ -1 +1,2 @@ {"effective_from": null, "last_verified_at": "2026-07-05T00:00:00Z", "model_id": "gpt-5", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", "official_source_url": "https://platform.openai.com/docs/pricing", "pricing": {"batch_input": 0.625, "batch_output": 5.0, "cache_write": null, "cached_input": 0.125, "currency": "USD", "input": 1.25, "output": 10.0, "unit": "1M tokens"}, "provider_id": "openai", "recorded_at": "2026-07-05T00:00:00Z"} +{"effective_from": null, "last_verified_at": "2026-07-27T00:00:00Z", "model_id": "gpt-5", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", "official_source_url": "https://platform.openai.com/docs/pricing", "pricing": {"batch_input": 0.625, "batch_output": 5.0, "cache_write": null, "cached_input": 0.125, "currency": "USD", "input": 1.25, "output": 10.0, "unit": "1M tokens"}, "provider_id": "openai", "recorded_at": "2026-07-27T00:00:00Z"} diff --git a/data/history/openai/o3.jsonl b/data/history/openai/o3.jsonl index dae584c..29b962b 100644 --- a/data/history/openai/o3.jsonl +++ b/data/history/openai/o3.jsonl @@ -1 +1,2 @@ {"effective_from": null, "last_verified_at": "2026-07-05T00:00:00Z", "model_id": "o3", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", "official_source_url": "https://platform.openai.com/docs/pricing", "pricing": {"batch_input": 1.0, "batch_output": 4.0, "cache_write": null, "cached_input": 0.5, "currency": "USD", "input": 2.0, "output": 8.0, "unit": "1M tokens"}, "provider_id": "openai", "recorded_at": "2026-07-05T00:00:00Z"} +{"effective_from": null, "last_verified_at": "2026-07-27T00:00:00Z", "model_id": "o3", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", "official_source_url": "https://platform.openai.com/docs/pricing", "pricing": {"batch_input": 1.0, "batch_output": 4.0, "cache_write": null, "cached_input": 0.5, "currency": "USD", "input": 2.0, "output": 8.0, "unit": "1M tokens"}, "provider_id": "openai", "recorded_at": "2026-07-27T00:00:00Z"} diff --git a/data/history/xai/grok-4.3.jsonl b/data/history/xai/grok-4.3.jsonl index 3506e08..6ca19e4 100644 --- a/data/history/xai/grok-4.3.jsonl +++ b/data/history/xai/grok-4.3.jsonl @@ -1 +1,2 @@ {"effective_from": null, "last_verified_at": "2026-07-05T00:00:00Z", "model_id": "grok-4.3", "notes": "xAI official model page lists Grok 4.3 input and output rates per 1M tokens. Cached and batch rates were not shown in the extracted official model card.", "official_source_url": "https://docs.x.ai/developers/models", "pricing": {"batch_input": null, "batch_output": null, "cache_write": null, "cached_input": null, "currency": "USD", "input": 1.25, "output": 2.5, "unit": "1M tokens"}, "provider_id": "xai", "recorded_at": "2026-07-05T00:00:00Z"} +{"effective_from": null, "last_verified_at": "2026-07-27T00:00:00Z", "model_id": "grok-4.3", "notes": "V1 stores short-context pricing. The long-context threshold is 200K tokens. Grok 4.3 batch pricing receives a 20% discount. Long-context prices are not stored in the current short-context fields.", "official_source_url": "https://docs.x.ai/developers/pricing", "pricing": {"batch_input": 1.0, "batch_output": 2.0, "cache_write": null, "cached_input": 0.2, "currency": "USD", "input": 1.25, "output": 2.5, "unit": "1M tokens"}, "provider_id": "xai", "recorded_at": "2026-07-27T00:00:00Z"} diff --git a/data/models/anthropic/claude-haiku-4.5.json b/data/models/anthropic/claude-haiku-4.5.json index 870db72..c201ead 100644 --- a/data/models/anthropic/claude-haiku-4.5.json +++ b/data/models/anthropic/claude-haiku-4.5.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Claude Haiku 4.5", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Claude Haiku", "model_id": "claude-haiku-4.5", "notes": "cache_write stores the 5-minute write rate.", diff --git a/data/models/anthropic/claude-opus-4.8.json b/data/models/anthropic/claude-opus-4.8.json index 2ee11c5..d5e2db9 100644 --- a/data/models/anthropic/claude-opus-4.8.json +++ b/data/models/anthropic/claude-opus-4.8.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Claude Opus 4.8", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Claude Opus", "model_id": "claude-opus-4.8", "notes": "Anthropic table columns are input, 5-minute cache write, 1-hour cache write, cache read, and output. cache_write stores the 5-minute write rate.", diff --git a/data/models/anthropic/claude-sonnet-4.6.json b/data/models/anthropic/claude-sonnet-4.6.json index 8e566c8..e9a5590 100644 --- a/data/models/anthropic/claude-sonnet-4.6.json +++ b/data/models/anthropic/claude-sonnet-4.6.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Claude Sonnet 4.6", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Claude Sonnet", "model_id": "claude-sonnet-4.6", "notes": "cache_write stores the 5-minute write rate.", diff --git a/data/models/anthropic/claude-sonnet-5-intro.json b/data/models/anthropic/claude-sonnet-5-intro.json index a24f635..af4499a 100644 --- a/data/models/anthropic/claude-sonnet-5-intro.json +++ b/data/models/anthropic/claude-sonnet-5-intro.json @@ -1,12 +1,12 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Claude Sonnet 5 introductory pricing", - "effective_from": "2026-07-05", - "last_verified_at": "2026-07-05T00:00:00Z", + "effective_from": "2026-06-30", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Claude Sonnet", "model_id": "claude-sonnet-5-intro", - "notes": "Introductory pricing is listed by Anthropic as effective through 2026-08-31; standard pricing starts 2026-09-01.", - "official_source_url": "https://docs.anthropic.com/en/docs/about-claude/pricing#claude-sonnet-5-introductory-pricing", + "notes": "Introductory pricing applies through 2026-08-31 and standard pricing starts 2026-09-01. V1 must not switch early to the standard $3/M input and $15/M output rates.", + "official_source_url": "https://www.anthropic.com/news/claude-sonnet-5", "pricing": { "batch_input": 1.0, "batch_output": 5.0, diff --git a/data/models/cohere/aya-expanse-32b.json b/data/models/cohere/aya-expanse-32b.json index f414dbd..f8ba278 100644 --- a/data/models/cohere/aya-expanse-32b.json +++ b/data/models/cohere/aya-expanse-32b.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Aya Expanse 32B", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Aya", "model_id": "aya-expanse-32b", "notes": "Cohere official pricing page states Aya Expanse 8B and 32B API pricing is $0.50/1M input and $1.50/1M output.", diff --git a/data/models/cohere/command-a-plus.json b/data/models/cohere/command-a-plus.json index 3e70177..0d4f45b 100644 --- a/data/models/cohere/command-a-plus.json +++ b/data/models/cohere/command-a-plus.json @@ -1,12 +1,12 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Command A+", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Command", "model_id": "command-a-plus", - "notes": "Cohere official pricing page lists Command A+ but the extracted public page did not expose a confirmed token price. Values are null rather than guessed.", - "official_source_url": "https://cohere.com/pricing", + "notes": "Cohere's official Command A+ page states that access is free until applicable rate limits are reached and that production use is available through Model Vault. Another official Cohere page shows per-token pricing for the same model ID, so V1 keeps token pricing null until the official documentation is reconciled.", + "official_source_url": "https://docs.cohere.com/docs/command-a-plus", "pricing": { "batch_input": null, "batch_output": null, diff --git a/data/models/cohere/command-r-plus-08-2024.json b/data/models/cohere/command-r-plus-08-2024.json index f9e6f83..d10294b 100644 --- a/data/models/cohere/command-r-plus-08-2024.json +++ b/data/models/cohere/command-r-plus-08-2024.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Command R+ 08-2024", "effective_from": "2024-08-01", - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Command R", "model_id": "command-r-plus-08-2024", "notes": "Cohere official pricing page lists Command R+ 08-2024 at $2.50/1M input tokens and $10.00/1M output tokens.", diff --git a/data/models/deepseek/deepseek-v4-flash.json b/data/models/deepseek/deepseek-v4-flash.json index a85837f..52f31ad 100644 --- a/data/models/deepseek/deepseek-v4-flash.json +++ b/data/models/deepseek/deepseek-v4-flash.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "DeepSeek V4 Flash", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "DeepSeek V4", "model_id": "deepseek-v4-flash", "notes": "DeepSeek labels input cache miss as input price and cache hit as cached_input.", diff --git a/data/models/deepseek/deepseek-v4-pro.json b/data/models/deepseek/deepseek-v4-pro.json index e5f98bc..1d6a1f6 100644 --- a/data/models/deepseek/deepseek-v4-pro.json +++ b/data/models/deepseek/deepseek-v4-pro.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "DeepSeek V4 Pro", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "DeepSeek V4", "model_id": "deepseek-v4-pro", "notes": "DeepSeek labels input cache miss as input price and cache hit as cached_input.", diff --git a/data/models/google-gemini/gemini-2.5-flash.json b/data/models/google-gemini/gemini-2.5-flash.json index 69bbec5..91bea1e 100644 --- a/data/models/google-gemini/gemini-2.5-flash.json +++ b/data/models/google-gemini/gemini-2.5-flash.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Gemini 2.5 Flash", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Gemini 2.5", "model_id": "gemini-2.5-flash", "notes": "Rates are for text/image/video input; Google lists separate audio rates in the official pricing table.", diff --git a/data/models/google-gemini/gemini-2.5-pro.json b/data/models/google-gemini/gemini-2.5-pro.json index fb63bed..32997b2 100644 --- a/data/models/google-gemini/gemini-2.5-pro.json +++ b/data/models/google-gemini/gemini-2.5-pro.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Gemini 2.5 Pro", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Gemini 2.5", "model_id": "gemini-2.5-pro", "notes": "Rates are for prompts up to 200k tokens. Google lists higher rates above 200k tokens in the official pricing table.", diff --git a/data/models/mistral-ai/mistral-large.json b/data/models/mistral-ai/mistral-large.json index 4b20e7b..109a096 100644 --- a/data/models/mistral-ai/mistral-large.json +++ b/data/models/mistral-ai/mistral-large.json @@ -1,20 +1,20 @@ { - "accessed_at": "2026-07-05T00:00:00Z", - "display_name": "Mistral Large", + "accessed_at": "2026-07-27T00:00:00Z", + "display_name": "Mistral Large 3", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Mistral Large", "model_id": "mistral-large", - "notes": "Mistral official pricing FAQ states Mistral Large costs $2/M input and $6/M output and batch processing gets a 50% discount.", - "official_source_url": "https://mistral.ai/pricing", + "notes": "mistral-large-latest currently maps to Mistral Large 3. Standard price is $0.50/M input and $1.50/M output. Cached input receives a 90% discount. Batch processing receives a 50% discount.", + "official_source_url": "https://mistral.ai/pricing/api/", "pricing": { - "batch_input": 1.0, - "batch_output": 3.0, + "batch_input": 0.25, + "batch_output": 0.75, "cache_write": null, - "cached_input": null, + "cached_input": 0.05, "currency": "USD", - "input": 2.0, - "output": 6.0, + "input": 0.5, + "output": 1.5, "unit": "1M tokens" }, "provider_id": "mistral-ai", diff --git a/data/models/openai/gpt-4.1.json b/data/models/openai/gpt-4.1.json index c2da567..6bc6e29 100644 --- a/data/models/openai/gpt-4.1.json +++ b/data/models/openai/gpt-4.1.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-4.1", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-4.1", "model_id": "gpt-4.1", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", diff --git a/data/models/openai/gpt-5.4-mini.json b/data/models/openai/gpt-5.4-mini.json index a46ae5f..0207b01 100644 --- a/data/models/openai/gpt-5.4-mini.json +++ b/data/models/openai/gpt-5.4-mini.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5.4 mini", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5.4", "model_id": "gpt-5.4-mini", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", diff --git a/data/models/openai/gpt-5.5.json b/data/models/openai/gpt-5.5.json index 39e1e09..07f4d09 100644 --- a/data/models/openai/gpt-5.5.json +++ b/data/models/openai/gpt-5.5.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5.5", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5.5", "model_id": "gpt-5.5", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens. Batch prices are 50% of the listed standard price where shown.", diff --git a/data/models/openai/gpt-5.6-luna.json b/data/models/openai/gpt-5.6-luna.json index 515bccf..12aecbb 100644 --- a/data/models/openai/gpt-5.6-luna.json +++ b/data/models/openai/gpt-5.6-luna.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-10T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5.6 Luna", "effective_from": null, - "last_verified_at": "2026-07-10T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5.6", "model_id": "gpt-5.6-luna", "notes": "OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices.", diff --git a/data/models/openai/gpt-5.6-sol.json b/data/models/openai/gpt-5.6-sol.json index 8cb9bfa..8fe085e 100644 --- a/data/models/openai/gpt-5.6-sol.json +++ b/data/models/openai/gpt-5.6-sol.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-10T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5.6 Sol", "effective_from": null, - "last_verified_at": "2026-07-10T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5.6", "model_id": "gpt-5.6-sol", "notes": "OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices.", diff --git a/data/models/openai/gpt-5.6-terra.json b/data/models/openai/gpt-5.6-terra.json index 1c0fa46..3531a31 100644 --- a/data/models/openai/gpt-5.6-terra.json +++ b/data/models/openai/gpt-5.6-terra.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-10T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5.6 Terra", "effective_from": null, - "last_verified_at": "2026-07-10T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5.6", "model_id": "gpt-5.6-terra", "notes": "OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices.", diff --git a/data/models/openai/gpt-5.json b/data/models/openai/gpt-5.json index 76badf4..1ba3653 100644 --- a/data/models/openai/gpt-5.json +++ b/data/models/openai/gpt-5.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5", "model_id": "gpt-5", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", diff --git a/data/models/openai/o3.json b/data/models/openai/o3.json index a9aad76..4f9c4e3 100644 --- a/data/models/openai/o3.json +++ b/data/models/openai/o3.json @@ -1,8 +1,8 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "o3", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "o-series", "model_id": "o3", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", diff --git a/data/models/xai/grok-4.3.json b/data/models/xai/grok-4.3.json index f14c96d..f4d16b8 100644 --- a/data/models/xai/grok-4.3.json +++ b/data/models/xai/grok-4.3.json @@ -1,17 +1,17 @@ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Grok 4.3", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Grok 4", "model_id": "grok-4.3", - "notes": "xAI official model page lists Grok 4.3 input and output rates per 1M tokens. Cached and batch rates were not shown in the extracted official model card.", - "official_source_url": "https://docs.x.ai/developers/models", + "notes": "V1 stores short-context pricing. The long-context threshold is 200K tokens. Grok 4.3 batch pricing receives a 20% discount. Long-context prices are not stored in the current short-context fields.", + "official_source_url": "https://docs.x.ai/developers/pricing", "pricing": { - "batch_input": null, - "batch_output": null, + "batch_input": 1.0, + "batch_output": 2.0, "cache_write": null, - "cached_input": null, + "cached_input": 0.2, "currency": "USD", "input": 1.25, "output": 2.5, diff --git a/data/prices.csv b/data/prices.csv index e88a766..47b95df 100644 --- a/data/prices.csv +++ b/data/prices.csv @@ -1,22 +1,22 @@ provider_id,model_id,display_name,model_family,status,currency,unit,input,output,cached_input,cache_write,batch_input,batch_output,official_source_url,accessed_at,last_verified_at,effective_from,notes -anthropic,claude-haiku-4.5,Claude Haiku 4.5,Claude Haiku,active,USD,1M tokens,1.0,5.0,0.1,1.25,0.5,2.5,https://docs.anthropic.com/en/docs/about-claude/pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,,cache_write stores the 5-minute write rate. -anthropic,claude-opus-4.8,Claude Opus 4.8,Claude Opus,active,USD,1M tokens,5.0,25.0,0.5,6.25,2.5,12.5,https://docs.anthropic.com/en/docs/about-claude/pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,,"Anthropic table columns are input, 5-minute cache write, 1-hour cache write, cache read, and output. cache_write stores the 5-minute write rate." -anthropic,claude-sonnet-4.6,Claude Sonnet 4.6,Claude Sonnet,active,USD,1M tokens,3.0,15.0,0.3,3.75,1.5,7.5,https://docs.anthropic.com/en/docs/about-claude/pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,,cache_write stores the 5-minute write rate. -anthropic,claude-sonnet-5-intro,Claude Sonnet 5 introductory pricing,Claude Sonnet,active,USD,1M tokens,2.0,10.0,0.2,2.5,1.0,5.0,https://docs.anthropic.com/en/docs/about-claude/pricing#claude-sonnet-5-introductory-pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,2026-07-05,Introductory pricing is listed by Anthropic as effective through 2026-08-31; standard pricing starts 2026-09-01. -cohere,aya-expanse-32b,Aya Expanse 32B,Aya,active,USD,1M tokens,0.5,1.5,,,,,https://cohere.com/pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,,Cohere official pricing page states Aya Expanse 8B and 32B API pricing is $0.50/1M input and $1.50/1M output. -cohere,command-a-plus,Command A+,Command,active,USD,1M tokens,,,,,,,https://cohere.com/pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,,Cohere official pricing page lists Command A+ but the extracted public page did not expose a confirmed token price. Values are null rather than guessed. -cohere,command-r-plus-08-2024,Command R+ 08-2024,Command R,active,USD,1M tokens,2.5,10.0,,,,,https://cohere.com/pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,2024-08-01,Cohere official pricing page lists Command R+ 08-2024 at $2.50/1M input tokens and $10.00/1M output tokens. -deepseek,deepseek-v4-flash,DeepSeek V4 Flash,DeepSeek V4,active,USD,1M tokens,0.14,0.28,0.0028,,,,https://api-docs.deepseek.com/quick_start/pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,,DeepSeek labels input cache miss as input price and cache hit as cached_input. -deepseek,deepseek-v4-pro,DeepSeek V4 Pro,DeepSeek V4,active,USD,1M tokens,0.435,0.87,0.003625,,,,https://api-docs.deepseek.com/quick_start/pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,,DeepSeek labels input cache miss as input price and cache hit as cached_input. -google-gemini,gemini-2.5-flash,Gemini 2.5 Flash,Gemini 2.5,active,USD,1M tokens,0.3,2.5,0.03,,0.15,1.25,https://ai.google.dev/gemini-api/docs/pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,,Rates are for text/image/video input; Google lists separate audio rates in the official pricing table. -google-gemini,gemini-2.5-pro,Gemini 2.5 Pro,Gemini 2.5,active,USD,1M tokens,1.25,10.0,0.125,,0.625,5.0,https://ai.google.dev/gemini-api/docs/pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,,Rates are for prompts up to 200k tokens. Google lists higher rates above 200k tokens in the official pricing table. -mistral-ai,mistral-large,Mistral Large,Mistral Large,active,USD,1M tokens,2.0,6.0,,,1.0,3.0,https://mistral.ai/pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,,Mistral official pricing FAQ states Mistral Large costs $2/M input and $6/M output and batch processing gets a 50% discount. -openai,gpt-4.1,GPT-4.1,GPT-4.1,active,USD,1M tokens,2.0,8.0,0.5,,1.0,4.0,https://platform.openai.com/docs/pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,,OpenAI official pricing page lists standard and batch rates per 1M tokens. -openai,gpt-5,GPT-5,GPT-5,active,USD,1M tokens,1.25,10.0,0.125,,0.625,5.0,https://platform.openai.com/docs/pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,,OpenAI official pricing page lists standard and batch rates per 1M tokens. -openai,gpt-5.4-mini,GPT-5.4 mini,GPT-5.4,active,USD,1M tokens,0.75,4.5,0.075,,0.375,2.25,https://platform.openai.com/docs/pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,,OpenAI official pricing page lists standard and batch rates per 1M tokens. -openai,gpt-5.5,GPT-5.5,GPT-5.5,active,USD,1M tokens,5.0,30.0,0.5,,2.5,15.0,https://platform.openai.com/docs/pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,,OpenAI official pricing page lists standard and batch rates per 1M tokens. Batch prices are 50% of the listed standard price where shown. -openai,gpt-5.6-luna,GPT-5.6 Luna,GPT-5.6,active,USD,1M tokens,1.0,6.0,0.1,1.25,0.5,3.0,https://developers.openai.com/api/docs/pricing,2026-07-10T00:00:00Z,2026-07-10T00:00:00Z,,"OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices." -openai,gpt-5.6-sol,GPT-5.6 Sol,GPT-5.6,active,USD,1M tokens,5.0,30.0,0.5,6.25,2.5,15.0,https://developers.openai.com/api/docs/pricing,2026-07-10T00:00:00Z,2026-07-10T00:00:00Z,,"OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices." -openai,gpt-5.6-terra,GPT-5.6 Terra,GPT-5.6,active,USD,1M tokens,2.5,15.0,0.25,3.125,1.25,7.5,https://developers.openai.com/api/docs/pricing,2026-07-10T00:00:00Z,2026-07-10T00:00:00Z,,"OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices." -openai,o3,o3,o-series,active,USD,1M tokens,2.0,8.0,0.5,,1.0,4.0,https://platform.openai.com/docs/pricing,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,,OpenAI official pricing page lists standard and batch rates per 1M tokens. -xai,grok-4.3,Grok 4.3,Grok 4,active,USD,1M tokens,1.25,2.5,,,,,https://docs.x.ai/developers/models,2026-07-05T00:00:00Z,2026-07-05T00:00:00Z,,xAI official model page lists Grok 4.3 input and output rates per 1M tokens. Cached and batch rates were not shown in the extracted official model card. +anthropic,claude-haiku-4.5,Claude Haiku 4.5,Claude Haiku,active,USD,1M tokens,1.0,5.0,0.1,1.25,0.5,2.5,https://docs.anthropic.com/en/docs/about-claude/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,cache_write stores the 5-minute write rate. +anthropic,claude-opus-4.8,Claude Opus 4.8,Claude Opus,active,USD,1M tokens,5.0,25.0,0.5,6.25,2.5,12.5,https://docs.anthropic.com/en/docs/about-claude/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,"Anthropic table columns are input, 5-minute cache write, 1-hour cache write, cache read, and output. cache_write stores the 5-minute write rate." +anthropic,claude-sonnet-4.6,Claude Sonnet 4.6,Claude Sonnet,active,USD,1M tokens,3.0,15.0,0.3,3.75,1.5,7.5,https://docs.anthropic.com/en/docs/about-claude/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,cache_write stores the 5-minute write rate. +anthropic,claude-sonnet-5-intro,Claude Sonnet 5 introductory pricing,Claude Sonnet,active,USD,1M tokens,2.0,10.0,0.2,2.5,1.0,5.0,https://www.anthropic.com/news/claude-sonnet-5,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,2026-06-30,Introductory pricing applies through 2026-08-31 and standard pricing starts 2026-09-01. V1 must not switch early to the standard $3/M input and $15/M output rates. +cohere,aya-expanse-32b,Aya Expanse 32B,Aya,active,USD,1M tokens,0.5,1.5,,,,,https://cohere.com/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,Cohere official pricing page states Aya Expanse 8B and 32B API pricing is $0.50/1M input and $1.50/1M output. +cohere,command-a-plus,Command A+,Command,active,USD,1M tokens,,,,,,,https://docs.cohere.com/docs/command-a-plus,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,"Cohere's official Command A+ page states that access is free until applicable rate limits are reached and that production use is available through Model Vault. Another official Cohere page shows per-token pricing for the same model ID, so V1 keeps token pricing null until the official documentation is reconciled." +cohere,command-r-plus-08-2024,Command R+ 08-2024,Command R,active,USD,1M tokens,2.5,10.0,,,,,https://cohere.com/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,2024-08-01,Cohere official pricing page lists Command R+ 08-2024 at $2.50/1M input tokens and $10.00/1M output tokens. +deepseek,deepseek-v4-flash,DeepSeek V4 Flash,DeepSeek V4,active,USD,1M tokens,0.14,0.28,0.0028,,,,https://api-docs.deepseek.com/quick_start/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,DeepSeek labels input cache miss as input price and cache hit as cached_input. +deepseek,deepseek-v4-pro,DeepSeek V4 Pro,DeepSeek V4,active,USD,1M tokens,0.435,0.87,0.003625,,,,https://api-docs.deepseek.com/quick_start/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,DeepSeek labels input cache miss as input price and cache hit as cached_input. +google-gemini,gemini-2.5-flash,Gemini 2.5 Flash,Gemini 2.5,active,USD,1M tokens,0.3,2.5,0.03,,0.15,1.25,https://ai.google.dev/gemini-api/docs/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,Rates are for text/image/video input; Google lists separate audio rates in the official pricing table. +google-gemini,gemini-2.5-pro,Gemini 2.5 Pro,Gemini 2.5,active,USD,1M tokens,1.25,10.0,0.125,,0.625,5.0,https://ai.google.dev/gemini-api/docs/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,Rates are for prompts up to 200k tokens. Google lists higher rates above 200k tokens in the official pricing table. +mistral-ai,mistral-large,Mistral Large 3,Mistral Large,active,USD,1M tokens,0.5,1.5,0.05,,0.25,0.75,https://mistral.ai/pricing/api/,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,mistral-large-latest currently maps to Mistral Large 3. Standard price is $0.50/M input and $1.50/M output. Cached input receives a 90% discount. Batch processing receives a 50% discount. +openai,gpt-4.1,GPT-4.1,GPT-4.1,active,USD,1M tokens,2.0,8.0,0.5,,1.0,4.0,https://platform.openai.com/docs/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,OpenAI official pricing page lists standard and batch rates per 1M tokens. +openai,gpt-5,GPT-5,GPT-5,active,USD,1M tokens,1.25,10.0,0.125,,0.625,5.0,https://platform.openai.com/docs/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,OpenAI official pricing page lists standard and batch rates per 1M tokens. +openai,gpt-5.4-mini,GPT-5.4 mini,GPT-5.4,active,USD,1M tokens,0.75,4.5,0.075,,0.375,2.25,https://platform.openai.com/docs/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,OpenAI official pricing page lists standard and batch rates per 1M tokens. +openai,gpt-5.5,GPT-5.5,GPT-5.5,active,USD,1M tokens,5.0,30.0,0.5,,2.5,15.0,https://platform.openai.com/docs/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,OpenAI official pricing page lists standard and batch rates per 1M tokens. Batch prices are 50% of the listed standard price where shown. +openai,gpt-5.6-luna,GPT-5.6 Luna,GPT-5.6,active,USD,1M tokens,1.0,6.0,0.1,1.25,0.5,3.0,https://developers.openai.com/api/docs/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,"OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices." +openai,gpt-5.6-sol,GPT-5.6 Sol,GPT-5.6,active,USD,1M tokens,5.0,30.0,0.5,6.25,2.5,15.0,https://developers.openai.com/api/docs/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,"OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices." +openai,gpt-5.6-terra,GPT-5.6 Terra,GPT-5.6,active,USD,1M tokens,2.5,15.0,0.25,3.125,1.25,7.5,https://developers.openai.com/api/docs/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,"OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices." +openai,o3,o3,o-series,active,USD,1M tokens,2.0,8.0,0.5,,1.0,4.0,https://platform.openai.com/docs/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,OpenAI official pricing page lists standard and batch rates per 1M tokens. +xai,grok-4.3,Grok 4.3,Grok 4,active,USD,1M tokens,1.25,2.5,0.2,,1.0,2.0,https://docs.x.ai/developers/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,V1 stores short-context pricing. The long-context threshold is 200K tokens. Grok 4.3 batch pricing receives a 20% discount. Long-context prices are not stored in the current short-context fields. diff --git a/data/prices.json b/data/prices.json index 3967948..21cf181 100644 --- a/data/prices.json +++ b/data/prices.json @@ -2,9 +2,9 @@ "dataset_name": "AICostBudget AI API Pricing Dataset", "dataset_version": "1.0.0", "description": "Open, machine-readable pricing data for LLM and AI APIs.", - "generated_at": "2026-07-09T18:36:17Z", + "generated_at": "2026-07-27T00:00:00Z", "homepage": "https://aicostbudget.com", - "last_verified_at": "2026-07-10T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "licenses": { "code": "MIT", "data": "CC BY 4.0" @@ -12,10 +12,10 @@ "model_count": 21, "models": [ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Claude Haiku 4.5", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Claude Haiku", "model_id": "claude-haiku-4.5", "notes": "cache_write stores the 5-minute write rate.", @@ -34,10 +34,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Claude Opus 4.8", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Claude Opus", "model_id": "claude-opus-4.8", "notes": "Anthropic table columns are input, 5-minute cache write, 1-hour cache write, cache read, and output. cache_write stores the 5-minute write rate.", @@ -56,10 +56,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Claude Sonnet 4.6", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Claude Sonnet", "model_id": "claude-sonnet-4.6", "notes": "cache_write stores the 5-minute write rate.", @@ -78,14 +78,14 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Claude Sonnet 5 introductory pricing", - "effective_from": "2026-07-05", - "last_verified_at": "2026-07-05T00:00:00Z", + "effective_from": "2026-06-30", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Claude Sonnet", "model_id": "claude-sonnet-5-intro", - "notes": "Introductory pricing is listed by Anthropic as effective through 2026-08-31; standard pricing starts 2026-09-01.", - "official_source_url": "https://docs.anthropic.com/en/docs/about-claude/pricing#claude-sonnet-5-introductory-pricing", + "notes": "Introductory pricing applies through 2026-08-31 and standard pricing starts 2026-09-01. V1 must not switch early to the standard $3/M input and $15/M output rates.", + "official_source_url": "https://www.anthropic.com/news/claude-sonnet-5", "pricing": { "batch_input": 1.0, "batch_output": 5.0, @@ -100,10 +100,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Aya Expanse 32B", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Aya", "model_id": "aya-expanse-32b", "notes": "Cohere official pricing page states Aya Expanse 8B and 32B API pricing is $0.50/1M input and $1.50/1M output.", @@ -122,14 +122,14 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Command A+", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Command", "model_id": "command-a-plus", - "notes": "Cohere official pricing page lists Command A+ but the extracted public page did not expose a confirmed token price. Values are null rather than guessed.", - "official_source_url": "https://cohere.com/pricing", + "notes": "Cohere's official Command A+ page states that access is free until applicable rate limits are reached and that production use is available through Model Vault. Another official Cohere page shows per-token pricing for the same model ID, so V1 keeps token pricing null until the official documentation is reconciled.", + "official_source_url": "https://docs.cohere.com/docs/command-a-plus", "pricing": { "batch_input": null, "batch_output": null, @@ -144,10 +144,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Command R+ 08-2024", "effective_from": "2024-08-01", - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Command R", "model_id": "command-r-plus-08-2024", "notes": "Cohere official pricing page lists Command R+ 08-2024 at $2.50/1M input tokens and $10.00/1M output tokens.", @@ -166,10 +166,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "DeepSeek V4 Flash", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "DeepSeek V4", "model_id": "deepseek-v4-flash", "notes": "DeepSeek labels input cache miss as input price and cache hit as cached_input.", @@ -188,10 +188,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "DeepSeek V4 Pro", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "DeepSeek V4", "model_id": "deepseek-v4-pro", "notes": "DeepSeek labels input cache miss as input price and cache hit as cached_input.", @@ -210,10 +210,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Gemini 2.5 Flash", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Gemini 2.5", "model_id": "gemini-2.5-flash", "notes": "Rates are for text/image/video input; Google lists separate audio rates in the official pricing table.", @@ -232,10 +232,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Gemini 2.5 Pro", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Gemini 2.5", "model_id": "gemini-2.5-pro", "notes": "Rates are for prompts up to 200k tokens. Google lists higher rates above 200k tokens in the official pricing table.", @@ -254,32 +254,32 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", - "display_name": "Mistral Large", + "accessed_at": "2026-07-27T00:00:00Z", + "display_name": "Mistral Large 3", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Mistral Large", "model_id": "mistral-large", - "notes": "Mistral official pricing FAQ states Mistral Large costs $2/M input and $6/M output and batch processing gets a 50% discount.", - "official_source_url": "https://mistral.ai/pricing", + "notes": "mistral-large-latest currently maps to Mistral Large 3. Standard price is $0.50/M input and $1.50/M output. Cached input receives a 90% discount. Batch processing receives a 50% discount.", + "official_source_url": "https://mistral.ai/pricing/api/", "pricing": { - "batch_input": 1.0, - "batch_output": 3.0, + "batch_input": 0.25, + "batch_output": 0.75, "cache_write": null, - "cached_input": null, + "cached_input": 0.05, "currency": "USD", - "input": 2.0, - "output": 6.0, + "input": 0.5, + "output": 1.5, "unit": "1M tokens" }, "provider_id": "mistral-ai", "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-4.1", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-4.1", "model_id": "gpt-4.1", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", @@ -298,10 +298,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5", "model_id": "gpt-5", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", @@ -320,10 +320,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5.4 mini", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5.4", "model_id": "gpt-5.4-mini", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", @@ -342,10 +342,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5.5", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5.5", "model_id": "gpt-5.5", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens. Batch prices are 50% of the listed standard price where shown.", @@ -364,10 +364,10 @@ "status": "active" }, { - "accessed_at": "2026-07-10T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5.6 Luna", "effective_from": null, - "last_verified_at": "2026-07-10T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5.6", "model_id": "gpt-5.6-luna", "notes": "OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices.", @@ -386,10 +386,10 @@ "status": "active" }, { - "accessed_at": "2026-07-10T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5.6 Sol", "effective_from": null, - "last_verified_at": "2026-07-10T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5.6", "model_id": "gpt-5.6-sol", "notes": "OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices.", @@ -408,10 +408,10 @@ "status": "active" }, { - "accessed_at": "2026-07-10T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5.6 Terra", "effective_from": null, - "last_verified_at": "2026-07-10T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5.6", "model_id": "gpt-5.6-terra", "notes": "OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices.", @@ -430,10 +430,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "o3", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "o-series", "model_id": "o3", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", @@ -452,19 +452,19 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Grok 4.3", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Grok 4", "model_id": "grok-4.3", - "notes": "xAI official model page lists Grok 4.3 input and output rates per 1M tokens. Cached and batch rates were not shown in the extracted official model card.", - "official_source_url": "https://docs.x.ai/developers/models", + "notes": "V1 stores short-context pricing. The long-context threshold is 200K tokens. Grok 4.3 batch pricing receives a 20% discount. Long-context prices are not stored in the current short-context fields.", + "official_source_url": "https://docs.x.ai/developers/pricing", "pricing": { - "batch_input": null, - "batch_output": null, + "batch_input": 1.0, + "batch_output": 2.0, "cache_write": null, - "cached_input": null, + "cached_input": 0.2, "currency": "USD", "input": 1.25, "output": 2.5, @@ -474,17 +474,18 @@ "status": "active" } ], - "official_source_count": 9, + "official_source_count": 10, "official_sources": [ "https://ai.google.dev/gemini-api/docs/pricing", "https://api-docs.deepseek.com/quick_start/pricing", "https://cohere.com/pricing", "https://developers.openai.com/api/docs/pricing", "https://docs.anthropic.com/en/docs/about-claude/pricing", - "https://docs.anthropic.com/en/docs/about-claude/pricing#claude-sonnet-5-introductory-pricing", - "https://docs.x.ai/developers/models", - "https://mistral.ai/pricing", - "https://platform.openai.com/docs/pricing" + "https://docs.cohere.com/docs/command-a-plus", + "https://docs.x.ai/developers/pricing", + "https://mistral.ai/pricing/api/", + "https://platform.openai.com/docs/pricing", + "https://www.anthropic.com/news/claude-sonnet-5" ], "provider_count": 7, "providers": [ diff --git a/data/providers/anthropic.json b/data/providers/anthropic.json index 41b2fa6..5e7be6c 100644 --- a/data/providers/anthropic.json +++ b/data/providers/anthropic.json @@ -3,10 +3,10 @@ "docs_url": "https://docs.anthropic.com", "models": [ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Claude Haiku 4.5", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Claude Haiku", "model_id": "claude-haiku-4.5", "notes": "cache_write stores the 5-minute write rate.", @@ -25,10 +25,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Claude Opus 4.8", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Claude Opus", "model_id": "claude-opus-4.8", "notes": "Anthropic table columns are input, 5-minute cache write, 1-hour cache write, cache read, and output. cache_write stores the 5-minute write rate.", @@ -47,10 +47,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Claude Sonnet 4.6", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Claude Sonnet", "model_id": "claude-sonnet-4.6", "notes": "cache_write stores the 5-minute write rate.", @@ -69,14 +69,14 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Claude Sonnet 5 introductory pricing", - "effective_from": "2026-07-05", - "last_verified_at": "2026-07-05T00:00:00Z", + "effective_from": "2026-06-30", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Claude Sonnet", "model_id": "claude-sonnet-5-intro", - "notes": "Introductory pricing is listed by Anthropic as effective through 2026-08-31; standard pricing starts 2026-09-01.", - "official_source_url": "https://docs.anthropic.com/en/docs/about-claude/pricing#claude-sonnet-5-introductory-pricing", + "notes": "Introductory pricing applies through 2026-08-31 and standard pricing starts 2026-09-01. V1 must not switch early to the standard $3/M input and $15/M output rates.", + "official_source_url": "https://www.anthropic.com/news/claude-sonnet-5", "pricing": { "batch_input": 1.0, "batch_output": 5.0, diff --git a/data/providers/cohere.json b/data/providers/cohere.json index d9b1bd9..c6cadff 100644 --- a/data/providers/cohere.json +++ b/data/providers/cohere.json @@ -3,10 +3,10 @@ "docs_url": "https://docs.cohere.com", "models": [ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Aya Expanse 32B", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Aya", "model_id": "aya-expanse-32b", "notes": "Cohere official pricing page states Aya Expanse 8B and 32B API pricing is $0.50/1M input and $1.50/1M output.", @@ -25,14 +25,14 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Command A+", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Command", "model_id": "command-a-plus", - "notes": "Cohere official pricing page lists Command A+ but the extracted public page did not expose a confirmed token price. Values are null rather than guessed.", - "official_source_url": "https://cohere.com/pricing", + "notes": "Cohere's official Command A+ page states that access is free until applicable rate limits are reached and that production use is available through Model Vault. Another official Cohere page shows per-token pricing for the same model ID, so V1 keeps token pricing null until the official documentation is reconciled.", + "official_source_url": "https://docs.cohere.com/docs/command-a-plus", "pricing": { "batch_input": null, "batch_output": null, @@ -47,10 +47,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Command R+ 08-2024", "effective_from": "2024-08-01", - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Command R", "model_id": "command-r-plus-08-2024", "notes": "Cohere official pricing page lists Command R+ 08-2024 at $2.50/1M input tokens and $10.00/1M output tokens.", diff --git a/data/providers/deepseek.json b/data/providers/deepseek.json index b34bff6..16bd173 100644 --- a/data/providers/deepseek.json +++ b/data/providers/deepseek.json @@ -3,10 +3,10 @@ "docs_url": "https://api-docs.deepseek.com", "models": [ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "DeepSeek V4 Flash", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "DeepSeek V4", "model_id": "deepseek-v4-flash", "notes": "DeepSeek labels input cache miss as input price and cache hit as cached_input.", @@ -25,10 +25,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "DeepSeek V4 Pro", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "DeepSeek V4", "model_id": "deepseek-v4-pro", "notes": "DeepSeek labels input cache miss as input price and cache hit as cached_input.", diff --git a/data/providers/google-gemini.json b/data/providers/google-gemini.json index 8a77e2c..e28af5a 100644 --- a/data/providers/google-gemini.json +++ b/data/providers/google-gemini.json @@ -3,10 +3,10 @@ "docs_url": "https://ai.google.dev/gemini-api/docs", "models": [ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Gemini 2.5 Flash", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Gemini 2.5", "model_id": "gemini-2.5-flash", "notes": "Rates are for text/image/video input; Google lists separate audio rates in the official pricing table.", @@ -25,10 +25,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Gemini 2.5 Pro", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Gemini 2.5", "model_id": "gemini-2.5-pro", "notes": "Rates are for prompts up to 200k tokens. Google lists higher rates above 200k tokens in the official pricing table.", diff --git a/data/providers/mistral-ai.json b/data/providers/mistral-ai.json index de8792f..237f155 100644 --- a/data/providers/mistral-ai.json +++ b/data/providers/mistral-ai.json @@ -3,22 +3,22 @@ "docs_url": "https://docs.mistral.ai", "models": [ { - "accessed_at": "2026-07-05T00:00:00Z", - "display_name": "Mistral Large", + "accessed_at": "2026-07-27T00:00:00Z", + "display_name": "Mistral Large 3", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Mistral Large", "model_id": "mistral-large", - "notes": "Mistral official pricing FAQ states Mistral Large costs $2/M input and $6/M output and batch processing gets a 50% discount.", - "official_source_url": "https://mistral.ai/pricing", + "notes": "mistral-large-latest currently maps to Mistral Large 3. Standard price is $0.50/M input and $1.50/M output. Cached input receives a 90% discount. Batch processing receives a 50% discount.", + "official_source_url": "https://mistral.ai/pricing/api/", "pricing": { - "batch_input": 1.0, - "batch_output": 3.0, + "batch_input": 0.25, + "batch_output": 0.75, "cache_write": null, - "cached_input": null, + "cached_input": 0.05, "currency": "USD", - "input": 2.0, - "output": 6.0, + "input": 0.5, + "output": 1.5, "unit": "1M tokens" }, "provider_id": "mistral-ai", diff --git a/data/providers/openai.json b/data/providers/openai.json index cd5be73..e9283aa 100644 --- a/data/providers/openai.json +++ b/data/providers/openai.json @@ -3,10 +3,10 @@ "docs_url": "https://platform.openai.com/docs", "models": [ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-4.1", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-4.1", "model_id": "gpt-4.1", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", @@ -25,10 +25,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5", "model_id": "gpt-5", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", @@ -47,10 +47,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5.4 mini", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5.4", "model_id": "gpt-5.4-mini", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", @@ -69,10 +69,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5.5", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5.5", "model_id": "gpt-5.5", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens. Batch prices are 50% of the listed standard price where shown.", @@ -91,10 +91,10 @@ "status": "active" }, { - "accessed_at": "2026-07-10T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5.6 Luna", "effective_from": null, - "last_verified_at": "2026-07-10T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5.6", "model_id": "gpt-5.6-luna", "notes": "OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices.", @@ -113,10 +113,10 @@ "status": "active" }, { - "accessed_at": "2026-07-10T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5.6 Sol", "effective_from": null, - "last_verified_at": "2026-07-10T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5.6", "model_id": "gpt-5.6-sol", "notes": "OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices.", @@ -135,10 +135,10 @@ "status": "active" }, { - "accessed_at": "2026-07-10T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "GPT-5.6 Terra", "effective_from": null, - "last_verified_at": "2026-07-10T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "GPT-5.6", "model_id": "gpt-5.6-terra", "notes": "OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices.", @@ -157,10 +157,10 @@ "status": "active" }, { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "o3", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "o-series", "model_id": "o3", "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", diff --git a/data/providers/xai.json b/data/providers/xai.json index aa642d2..7780370 100644 --- a/data/providers/xai.json +++ b/data/providers/xai.json @@ -3,19 +3,19 @@ "docs_url": "https://docs.x.ai", "models": [ { - "accessed_at": "2026-07-05T00:00:00Z", + "accessed_at": "2026-07-27T00:00:00Z", "display_name": "Grok 4.3", "effective_from": null, - "last_verified_at": "2026-07-05T00:00:00Z", + "last_verified_at": "2026-07-27T00:00:00Z", "model_family": "Grok 4", "model_id": "grok-4.3", - "notes": "xAI official model page lists Grok 4.3 input and output rates per 1M tokens. Cached and batch rates were not shown in the extracted official model card.", - "official_source_url": "https://docs.x.ai/developers/models", + "notes": "V1 stores short-context pricing. The long-context threshold is 200K tokens. Grok 4.3 batch pricing receives a 20% discount. Long-context prices are not stored in the current short-context fields.", + "official_source_url": "https://docs.x.ai/developers/pricing", "pricing": { - "batch_input": null, - "batch_output": null, + "batch_input": 1.0, + "batch_output": 2.0, "cache_write": null, - "cached_input": null, + "cached_input": 0.2, "currency": "USD", "input": 1.25, "output": 2.5, diff --git a/data/snapshots/2026-07-27/prices.csv b/data/snapshots/2026-07-27/prices.csv new file mode 100644 index 0000000..47b95df --- /dev/null +++ b/data/snapshots/2026-07-27/prices.csv @@ -0,0 +1,22 @@ +provider_id,model_id,display_name,model_family,status,currency,unit,input,output,cached_input,cache_write,batch_input,batch_output,official_source_url,accessed_at,last_verified_at,effective_from,notes +anthropic,claude-haiku-4.5,Claude Haiku 4.5,Claude Haiku,active,USD,1M tokens,1.0,5.0,0.1,1.25,0.5,2.5,https://docs.anthropic.com/en/docs/about-claude/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,cache_write stores the 5-minute write rate. +anthropic,claude-opus-4.8,Claude Opus 4.8,Claude Opus,active,USD,1M tokens,5.0,25.0,0.5,6.25,2.5,12.5,https://docs.anthropic.com/en/docs/about-claude/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,"Anthropic table columns are input, 5-minute cache write, 1-hour cache write, cache read, and output. cache_write stores the 5-minute write rate." +anthropic,claude-sonnet-4.6,Claude Sonnet 4.6,Claude Sonnet,active,USD,1M tokens,3.0,15.0,0.3,3.75,1.5,7.5,https://docs.anthropic.com/en/docs/about-claude/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,cache_write stores the 5-minute write rate. +anthropic,claude-sonnet-5-intro,Claude Sonnet 5 introductory pricing,Claude Sonnet,active,USD,1M tokens,2.0,10.0,0.2,2.5,1.0,5.0,https://www.anthropic.com/news/claude-sonnet-5,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,2026-06-30,Introductory pricing applies through 2026-08-31 and standard pricing starts 2026-09-01. V1 must not switch early to the standard $3/M input and $15/M output rates. +cohere,aya-expanse-32b,Aya Expanse 32B,Aya,active,USD,1M tokens,0.5,1.5,,,,,https://cohere.com/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,Cohere official pricing page states Aya Expanse 8B and 32B API pricing is $0.50/1M input and $1.50/1M output. +cohere,command-a-plus,Command A+,Command,active,USD,1M tokens,,,,,,,https://docs.cohere.com/docs/command-a-plus,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,"Cohere's official Command A+ page states that access is free until applicable rate limits are reached and that production use is available through Model Vault. Another official Cohere page shows per-token pricing for the same model ID, so V1 keeps token pricing null until the official documentation is reconciled." +cohere,command-r-plus-08-2024,Command R+ 08-2024,Command R,active,USD,1M tokens,2.5,10.0,,,,,https://cohere.com/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,2024-08-01,Cohere official pricing page lists Command R+ 08-2024 at $2.50/1M input tokens and $10.00/1M output tokens. +deepseek,deepseek-v4-flash,DeepSeek V4 Flash,DeepSeek V4,active,USD,1M tokens,0.14,0.28,0.0028,,,,https://api-docs.deepseek.com/quick_start/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,DeepSeek labels input cache miss as input price and cache hit as cached_input. +deepseek,deepseek-v4-pro,DeepSeek V4 Pro,DeepSeek V4,active,USD,1M tokens,0.435,0.87,0.003625,,,,https://api-docs.deepseek.com/quick_start/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,DeepSeek labels input cache miss as input price and cache hit as cached_input. +google-gemini,gemini-2.5-flash,Gemini 2.5 Flash,Gemini 2.5,active,USD,1M tokens,0.3,2.5,0.03,,0.15,1.25,https://ai.google.dev/gemini-api/docs/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,Rates are for text/image/video input; Google lists separate audio rates in the official pricing table. +google-gemini,gemini-2.5-pro,Gemini 2.5 Pro,Gemini 2.5,active,USD,1M tokens,1.25,10.0,0.125,,0.625,5.0,https://ai.google.dev/gemini-api/docs/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,Rates are for prompts up to 200k tokens. Google lists higher rates above 200k tokens in the official pricing table. +mistral-ai,mistral-large,Mistral Large 3,Mistral Large,active,USD,1M tokens,0.5,1.5,0.05,,0.25,0.75,https://mistral.ai/pricing/api/,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,mistral-large-latest currently maps to Mistral Large 3. Standard price is $0.50/M input and $1.50/M output. Cached input receives a 90% discount. Batch processing receives a 50% discount. +openai,gpt-4.1,GPT-4.1,GPT-4.1,active,USD,1M tokens,2.0,8.0,0.5,,1.0,4.0,https://platform.openai.com/docs/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,OpenAI official pricing page lists standard and batch rates per 1M tokens. +openai,gpt-5,GPT-5,GPT-5,active,USD,1M tokens,1.25,10.0,0.125,,0.625,5.0,https://platform.openai.com/docs/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,OpenAI official pricing page lists standard and batch rates per 1M tokens. +openai,gpt-5.4-mini,GPT-5.4 mini,GPT-5.4,active,USD,1M tokens,0.75,4.5,0.075,,0.375,2.25,https://platform.openai.com/docs/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,OpenAI official pricing page lists standard and batch rates per 1M tokens. +openai,gpt-5.5,GPT-5.5,GPT-5.5,active,USD,1M tokens,5.0,30.0,0.5,,2.5,15.0,https://platform.openai.com/docs/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,OpenAI official pricing page lists standard and batch rates per 1M tokens. Batch prices are 50% of the listed standard price where shown. +openai,gpt-5.6-luna,GPT-5.6 Luna,GPT-5.6,active,USD,1M tokens,1.0,6.0,0.1,1.25,0.5,3.0,https://developers.openai.com/api/docs/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,"OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices." +openai,gpt-5.6-sol,GPT-5.6 Sol,GPT-5.6,active,USD,1M tokens,5.0,30.0,0.5,6.25,2.5,15.0,https://developers.openai.com/api/docs/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,"OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices." +openai,gpt-5.6-terra,GPT-5.6 Terra,GPT-5.6,active,USD,1M tokens,2.5,15.0,0.25,3.125,1.25,7.5,https://developers.openai.com/api/docs/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,"OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices." +openai,o3,o3,o-series,active,USD,1M tokens,2.0,8.0,0.5,,1.0,4.0,https://platform.openai.com/docs/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,OpenAI official pricing page lists standard and batch rates per 1M tokens. +xai,grok-4.3,Grok 4.3,Grok 4,active,USD,1M tokens,1.25,2.5,0.2,,1.0,2.0,https://docs.x.ai/developers/pricing,2026-07-27T00:00:00Z,2026-07-27T00:00:00Z,,V1 stores short-context pricing. The long-context threshold is 200K tokens. Grok 4.3 batch pricing receives a 20% discount. Long-context prices are not stored in the current short-context fields. diff --git a/data/snapshots/2026-07-27/prices.json b/data/snapshots/2026-07-27/prices.json new file mode 100644 index 0000000..21cf181 --- /dev/null +++ b/data/snapshots/2026-07-27/prices.json @@ -0,0 +1,549 @@ +{ + "dataset_name": "AICostBudget AI API Pricing Dataset", + "dataset_version": "1.0.0", + "description": "Open, machine-readable pricing data for LLM and AI APIs.", + "generated_at": "2026-07-27T00:00:00Z", + "homepage": "https://aicostbudget.com", + "last_verified_at": "2026-07-27T00:00:00Z", + "licenses": { + "code": "MIT", + "data": "CC BY 4.0" + }, + "model_count": 21, + "models": [ + { + "accessed_at": "2026-07-27T00:00:00Z", + "display_name": "Claude Haiku 4.5", + "effective_from": null, + "last_verified_at": "2026-07-27T00:00:00Z", + "model_family": "Claude Haiku", + "model_id": "claude-haiku-4.5", + "notes": "cache_write stores the 5-minute write rate.", + "official_source_url": "https://docs.anthropic.com/en/docs/about-claude/pricing", + "pricing": { + "batch_input": 0.5, + "batch_output": 2.5, + "cache_write": 1.25, + "cached_input": 0.1, + "currency": "USD", + "input": 1.0, + "output": 5.0, + "unit": "1M tokens" + }, + "provider_id": "anthropic", + "status": "active" + }, + { + "accessed_at": "2026-07-27T00:00:00Z", + "display_name": "Claude Opus 4.8", + "effective_from": null, + "last_verified_at": "2026-07-27T00:00:00Z", + "model_family": "Claude Opus", + "model_id": "claude-opus-4.8", + "notes": "Anthropic table columns are input, 5-minute cache write, 1-hour cache write, cache read, and output. cache_write stores the 5-minute write rate.", + "official_source_url": "https://docs.anthropic.com/en/docs/about-claude/pricing", + "pricing": { + "batch_input": 2.5, + "batch_output": 12.5, + "cache_write": 6.25, + "cached_input": 0.5, + "currency": "USD", + "input": 5.0, + "output": 25.0, + "unit": "1M tokens" + }, + "provider_id": "anthropic", + "status": "active" + }, + { + "accessed_at": "2026-07-27T00:00:00Z", + "display_name": "Claude Sonnet 4.6", + "effective_from": null, + "last_verified_at": "2026-07-27T00:00:00Z", + "model_family": "Claude Sonnet", + "model_id": "claude-sonnet-4.6", + "notes": "cache_write stores the 5-minute write rate.", + "official_source_url": "https://docs.anthropic.com/en/docs/about-claude/pricing", + "pricing": { + "batch_input": 1.5, + "batch_output": 7.5, + "cache_write": 3.75, + "cached_input": 0.3, + "currency": "USD", + "input": 3.0, + "output": 15.0, + "unit": "1M tokens" + }, + "provider_id": "anthropic", + "status": "active" + }, + { + "accessed_at": "2026-07-27T00:00:00Z", + "display_name": "Claude Sonnet 5 introductory pricing", + "effective_from": "2026-06-30", + "last_verified_at": "2026-07-27T00:00:00Z", + "model_family": "Claude Sonnet", + "model_id": "claude-sonnet-5-intro", + "notes": "Introductory pricing applies through 2026-08-31 and standard pricing starts 2026-09-01. V1 must not switch early to the standard $3/M input and $15/M output rates.", + "official_source_url": "https://www.anthropic.com/news/claude-sonnet-5", + "pricing": { + "batch_input": 1.0, + "batch_output": 5.0, + "cache_write": 2.5, + "cached_input": 0.2, + "currency": "USD", + "input": 2.0, + "output": 10.0, + "unit": "1M tokens" + }, + "provider_id": "anthropic", + "status": "active" + }, + { + "accessed_at": "2026-07-27T00:00:00Z", + "display_name": "Aya Expanse 32B", + "effective_from": null, + "last_verified_at": "2026-07-27T00:00:00Z", + "model_family": "Aya", + "model_id": "aya-expanse-32b", + "notes": "Cohere official pricing page states Aya Expanse 8B and 32B API pricing is $0.50/1M input and $1.50/1M output.", + "official_source_url": "https://cohere.com/pricing", + "pricing": { + "batch_input": null, + "batch_output": null, + "cache_write": null, + "cached_input": null, + "currency": "USD", + "input": 0.5, + "output": 1.5, + "unit": "1M tokens" + }, + "provider_id": "cohere", + "status": "active" + }, + { + "accessed_at": "2026-07-27T00:00:00Z", + "display_name": "Command A+", + "effective_from": null, + "last_verified_at": "2026-07-27T00:00:00Z", + "model_family": "Command", + "model_id": "command-a-plus", + "notes": "Cohere's official Command A+ page states that access is free until applicable rate limits are reached and that production use is available through Model Vault. Another official Cohere page shows per-token pricing for the same model ID, so V1 keeps token pricing null until the official documentation is reconciled.", + "official_source_url": "https://docs.cohere.com/docs/command-a-plus", + "pricing": { + "batch_input": null, + "batch_output": null, + "cache_write": null, + "cached_input": null, + "currency": "USD", + "input": null, + "output": null, + "unit": "1M tokens" + }, + "provider_id": "cohere", + "status": "active" + }, + { + "accessed_at": "2026-07-27T00:00:00Z", + "display_name": "Command R+ 08-2024", + "effective_from": "2024-08-01", + "last_verified_at": "2026-07-27T00:00:00Z", + "model_family": "Command R", + "model_id": "command-r-plus-08-2024", + "notes": "Cohere official pricing page lists Command R+ 08-2024 at $2.50/1M input tokens and $10.00/1M output tokens.", + "official_source_url": "https://cohere.com/pricing", + "pricing": { + "batch_input": null, + "batch_output": null, + "cache_write": null, + "cached_input": null, + "currency": "USD", + "input": 2.5, + "output": 10.0, + "unit": "1M tokens" + }, + "provider_id": "cohere", + "status": "active" + }, + { + "accessed_at": "2026-07-27T00:00:00Z", + "display_name": "DeepSeek V4 Flash", + "effective_from": null, + "last_verified_at": "2026-07-27T00:00:00Z", + "model_family": "DeepSeek V4", + "model_id": "deepseek-v4-flash", + "notes": "DeepSeek labels input cache miss as input price and cache hit as cached_input.", + "official_source_url": "https://api-docs.deepseek.com/quick_start/pricing", + "pricing": { + "batch_input": null, + "batch_output": null, + "cache_write": null, + "cached_input": 0.0028, + "currency": "USD", + "input": 0.14, + "output": 0.28, + "unit": "1M tokens" + }, + "provider_id": "deepseek", + "status": "active" + }, + { + "accessed_at": "2026-07-27T00:00:00Z", + "display_name": "DeepSeek V4 Pro", + "effective_from": null, + "last_verified_at": "2026-07-27T00:00:00Z", + "model_family": "DeepSeek V4", + "model_id": "deepseek-v4-pro", + "notes": "DeepSeek labels input cache miss as input price and cache hit as cached_input.", + "official_source_url": "https://api-docs.deepseek.com/quick_start/pricing", + "pricing": { + "batch_input": null, + "batch_output": null, + "cache_write": null, + "cached_input": 0.003625, + "currency": "USD", + "input": 0.435, + "output": 0.87, + "unit": "1M tokens" + }, + "provider_id": "deepseek", + "status": "active" + }, + { + "accessed_at": "2026-07-27T00:00:00Z", + "display_name": "Gemini 2.5 Flash", + "effective_from": null, + "last_verified_at": "2026-07-27T00:00:00Z", + "model_family": "Gemini 2.5", + "model_id": "gemini-2.5-flash", + "notes": "Rates are for text/image/video input; Google lists separate audio rates in the official pricing table.", + "official_source_url": "https://ai.google.dev/gemini-api/docs/pricing", + "pricing": { + "batch_input": 0.15, + "batch_output": 1.25, + "cache_write": null, + "cached_input": 0.03, + "currency": "USD", + "input": 0.3, + "output": 2.5, + "unit": "1M tokens" + }, + "provider_id": "google-gemini", + "status": "active" + }, + { + "accessed_at": "2026-07-27T00:00:00Z", + "display_name": "Gemini 2.5 Pro", + "effective_from": null, + "last_verified_at": "2026-07-27T00:00:00Z", + "model_family": "Gemini 2.5", + "model_id": "gemini-2.5-pro", + "notes": "Rates are for prompts up to 200k tokens. Google lists higher rates above 200k tokens in the official pricing table.", + "official_source_url": "https://ai.google.dev/gemini-api/docs/pricing", + "pricing": { + "batch_input": 0.625, + "batch_output": 5.0, + "cache_write": null, + "cached_input": 0.125, + "currency": "USD", + "input": 1.25, + "output": 10.0, + "unit": "1M tokens" + }, + "provider_id": "google-gemini", + "status": "active" + }, + { + "accessed_at": "2026-07-27T00:00:00Z", + "display_name": "Mistral Large 3", + "effective_from": null, + "last_verified_at": "2026-07-27T00:00:00Z", + "model_family": "Mistral Large", + "model_id": "mistral-large", + "notes": "mistral-large-latest currently maps to Mistral Large 3. Standard price is $0.50/M input and $1.50/M output. Cached input receives a 90% discount. Batch processing receives a 50% discount.", + "official_source_url": "https://mistral.ai/pricing/api/", + "pricing": { + "batch_input": 0.25, + "batch_output": 0.75, + "cache_write": null, + "cached_input": 0.05, + "currency": "USD", + "input": 0.5, + "output": 1.5, + "unit": "1M tokens" + }, + "provider_id": "mistral-ai", + "status": "active" + }, + { + "accessed_at": "2026-07-27T00:00:00Z", + "display_name": "GPT-4.1", + "effective_from": null, + "last_verified_at": "2026-07-27T00:00:00Z", + "model_family": "GPT-4.1", + "model_id": "gpt-4.1", + "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", + "official_source_url": "https://platform.openai.com/docs/pricing", + "pricing": { + "batch_input": 1.0, + "batch_output": 4.0, + "cache_write": null, + "cached_input": 0.5, + "currency": "USD", + "input": 2.0, + "output": 8.0, + "unit": "1M tokens" + }, + "provider_id": "openai", + "status": "active" + }, + { + "accessed_at": "2026-07-27T00:00:00Z", + "display_name": "GPT-5", + "effective_from": null, + "last_verified_at": "2026-07-27T00:00:00Z", + "model_family": "GPT-5", + "model_id": "gpt-5", + "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", + "official_source_url": "https://platform.openai.com/docs/pricing", + "pricing": { + "batch_input": 0.625, + "batch_output": 5.0, + "cache_write": null, + "cached_input": 0.125, + "currency": "USD", + "input": 1.25, + "output": 10.0, + "unit": "1M tokens" + }, + "provider_id": "openai", + "status": "active" + }, + { + "accessed_at": "2026-07-27T00:00:00Z", + "display_name": "GPT-5.4 mini", + "effective_from": null, + "last_verified_at": "2026-07-27T00:00:00Z", + "model_family": "GPT-5.4", + "model_id": "gpt-5.4-mini", + "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", + "official_source_url": "https://platform.openai.com/docs/pricing", + "pricing": { + "batch_input": 0.375, + "batch_output": 2.25, + "cache_write": null, + "cached_input": 0.075, + "currency": "USD", + "input": 0.75, + "output": 4.5, + "unit": "1M tokens" + }, + "provider_id": "openai", + "status": "active" + }, + { + "accessed_at": "2026-07-27T00:00:00Z", + "display_name": "GPT-5.5", + "effective_from": null, + "last_verified_at": "2026-07-27T00:00:00Z", + "model_family": "GPT-5.5", + "model_id": "gpt-5.5", + "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens. Batch prices are 50% of the listed standard price where shown.", + "official_source_url": "https://platform.openai.com/docs/pricing", + "pricing": { + "batch_input": 2.5, + "batch_output": 15.0, + "cache_write": null, + "cached_input": 0.5, + "currency": "USD", + "input": 5.0, + "output": 30.0, + "unit": "1M tokens" + }, + "provider_id": "openai", + "status": "active" + }, + { + "accessed_at": "2026-07-27T00:00:00Z", + "display_name": "GPT-5.6 Luna", + "effective_from": null, + "last_verified_at": "2026-07-27T00:00:00Z", + "model_family": "GPT-5.6", + "model_id": "gpt-5.6-luna", + "notes": "OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices.", + "official_source_url": "https://developers.openai.com/api/docs/pricing", + "pricing": { + "batch_input": 0.5, + "batch_output": 3.0, + "cache_write": 1.25, + "cached_input": 0.1, + "currency": "USD", + "input": 1.0, + "output": 6.0, + "unit": "1M tokens" + }, + "provider_id": "openai", + "status": "active" + }, + { + "accessed_at": "2026-07-27T00:00:00Z", + "display_name": "GPT-5.6 Sol", + "effective_from": null, + "last_verified_at": "2026-07-27T00:00:00Z", + "model_family": "GPT-5.6", + "model_id": "gpt-5.6-sol", + "notes": "OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices.", + "official_source_url": "https://developers.openai.com/api/docs/pricing", + "pricing": { + "batch_input": 2.5, + "batch_output": 15.0, + "cache_write": 6.25, + "cached_input": 0.5, + "currency": "USD", + "input": 5.0, + "output": 30.0, + "unit": "1M tokens" + }, + "provider_id": "openai", + "status": "active" + }, + { + "accessed_at": "2026-07-27T00:00:00Z", + "display_name": "GPT-5.6 Terra", + "effective_from": null, + "last_verified_at": "2026-07-27T00:00:00Z", + "model_family": "GPT-5.6", + "model_id": "gpt-5.6-terra", + "notes": "OpenAI official pricing page lists Standard, Batch, Flex, and Priority rates per 1M tokens. Public V1 fields store Standard short-context prices and Batch short-context prices.", + "official_source_url": "https://developers.openai.com/api/docs/pricing", + "pricing": { + "batch_input": 1.25, + "batch_output": 7.5, + "cache_write": 3.125, + "cached_input": 0.25, + "currency": "USD", + "input": 2.5, + "output": 15.0, + "unit": "1M tokens" + }, + "provider_id": "openai", + "status": "active" + }, + { + "accessed_at": "2026-07-27T00:00:00Z", + "display_name": "o3", + "effective_from": null, + "last_verified_at": "2026-07-27T00:00:00Z", + "model_family": "o-series", + "model_id": "o3", + "notes": "OpenAI official pricing page lists standard and batch rates per 1M tokens.", + "official_source_url": "https://platform.openai.com/docs/pricing", + "pricing": { + "batch_input": 1.0, + "batch_output": 4.0, + "cache_write": null, + "cached_input": 0.5, + "currency": "USD", + "input": 2.0, + "output": 8.0, + "unit": "1M tokens" + }, + "provider_id": "openai", + "status": "active" + }, + { + "accessed_at": "2026-07-27T00:00:00Z", + "display_name": "Grok 4.3", + "effective_from": null, + "last_verified_at": "2026-07-27T00:00:00Z", + "model_family": "Grok 4", + "model_id": "grok-4.3", + "notes": "V1 stores short-context pricing. The long-context threshold is 200K tokens. Grok 4.3 batch pricing receives a 20% discount. Long-context prices are not stored in the current short-context fields.", + "official_source_url": "https://docs.x.ai/developers/pricing", + "pricing": { + "batch_input": 1.0, + "batch_output": 2.0, + "cache_write": null, + "cached_input": 0.2, + "currency": "USD", + "input": 1.25, + "output": 2.5, + "unit": "1M tokens" + }, + "provider_id": "xai", + "status": "active" + } + ], + "official_source_count": 10, + "official_sources": [ + "https://ai.google.dev/gemini-api/docs/pricing", + "https://api-docs.deepseek.com/quick_start/pricing", + "https://cohere.com/pricing", + "https://developers.openai.com/api/docs/pricing", + "https://docs.anthropic.com/en/docs/about-claude/pricing", + "https://docs.cohere.com/docs/command-a-plus", + "https://docs.x.ai/developers/pricing", + "https://mistral.ai/pricing/api/", + "https://platform.openai.com/docs/pricing", + "https://www.anthropic.com/news/claude-sonnet-5" + ], + "provider_count": 7, + "providers": [ + { + "display_name": "Anthropic", + "docs_url": "https://docs.anthropic.com", + "notes": "Official Claude API pricing documentation.", + "pricing_url": "https://docs.anthropic.com/en/docs/about-claude/pricing", + "provider_id": "anthropic", + "website_url": "https://www.anthropic.com" + }, + { + "display_name": "Cohere", + "docs_url": "https://docs.cohere.com", + "notes": "Official Cohere pricing page.", + "pricing_url": "https://cohere.com/pricing", + "provider_id": "cohere", + "website_url": "https://cohere.com" + }, + { + "display_name": "DeepSeek", + "docs_url": "https://api-docs.deepseek.com", + "notes": "Official DeepSeek API models and pricing documentation.", + "pricing_url": "https://api-docs.deepseek.com/quick_start/pricing", + "provider_id": "deepseek", + "website_url": "https://www.deepseek.com" + }, + { + "display_name": "Google Gemini", + "docs_url": "https://ai.google.dev/gemini-api/docs", + "notes": "Official Gemini API pricing documentation.", + "pricing_url": "https://ai.google.dev/gemini-api/docs/pricing", + "provider_id": "google-gemini", + "website_url": "https://ai.google.dev" + }, + { + "display_name": "Mistral AI", + "docs_url": "https://docs.mistral.ai", + "notes": "Official Mistral AI pricing page.", + "pricing_url": "https://mistral.ai/pricing", + "provider_id": "mistral-ai", + "website_url": "https://mistral.ai" + }, + { + "display_name": "OpenAI", + "docs_url": "https://platform.openai.com/docs", + "notes": "Official OpenAI API pricing documentation.", + "pricing_url": "https://platform.openai.com/docs/pricing", + "provider_id": "openai", + "website_url": "https://openai.com" + }, + { + "display_name": "xAI", + "docs_url": "https://docs.x.ai", + "notes": "Official xAI model documentation.", + "pricing_url": "https://docs.x.ai/developers/models", + "provider_id": "xai", + "website_url": "https://x.ai" + } + ] +}