agent: document Vercel, Mistral, llama.cpp providers and effort defaults#71
Merged
Conversation
- Add Vercel AI Gateway, Mistral, and llama.cpp to the providers table and the --provider allowed values / env-key lists - Extend the API-key auto-detection order with AI_GATEWAY_API_KEY and MISTRAL_API_KEY - Note llama.cpp as a probed local server (localhost:8080) alongside Ollama - Document HF_BILL_TO for Hugging Face org billing - Explain effort resolution (flag > remembered > provider default > mode default) and the Mistral none default
b7cc022 to
2a77cdd
Compare
krichprollsch
approved these changes
Jun 18, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Updates the
lightpanda agentdocs to match recent changes in thebrowserandzenairepos. The agent gained new LLM providers and effort behavior that the docs didn't reflect.Providers
--provider vercel, envAI_GATEWAY_API_KEY--provider mistral, envMISTRAL_API_KEY--provider llama_cpp, local (no key), basehttp://localhost:8080/v1, probed alongside OllamaNote the Vercel flag is
verceleven though its env var isAI_GATEWAY_API_KEY.Other changes
AI_GATEWAY_API_KEY→MISTRAL_API_KEYHF_BILL_TOfor billing Hugging Face routed requests to an orgnonebecause its default model rejects reasoning effortFiles
src/content/usage/agent.mdx— providers table, auto-detect order, local probing, HF billing, effortsrc/content/run-locally/commands/agent.mdx—--providerallowed values & env lists,--base-url,--effortdefaultsNote for maintainers
browser/src/help.zon(the source of the--helptext mirrored incommands/agent.mdx) is also stale — it still lists onlyanthropic, openai, gemini, huggingface, ollamaand omits the new env vars. These docs now reflect reality; a follow-up in the browser repo should refreshhelp.zonso--helpand the docs reconverge.Verification
Verified every flag / env var / default against the source:
zenai/src/provider.zig(openAiPreset,default_candidates,defaultEffort),browser/src/agent/settings.zig(api_keys_hint,resolveEffort), andbrowser/src/Config.zig(agent option set). Biome/typecheck can't run here (the package needs the parent monorepo workspace for@lightpanda/common); changes are pure markdown prose/tables.