Add routefoundry (Testing, Evaluation and Observability) - #165
Open
RitikPatill wants to merge 1 commit into
Open
Add routefoundry (Testing, Evaluation and Observability)#165RitikPatill wants to merge 1 commit into
RitikPatill wants to merge 1 commit into
Conversation
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Adds routefoundry, MIT-licensed.
It benchmarks the models you already have installed in Ollama using a 38-task auto-gradable suite graded deterministically (exact number / exact string / JSON field / regex), so there is no LLM judge whose bias would itself need auditing. It then audits whether routing between those models actually helps, and compiles an explainable policy.
Measured example from the repo, with raw rows committed: across 152 generations on a 16 GB laptop, no model won every category. deepseek-r1:1.5b (1.1 GB) answered 10/12 arithmetic word problems where llama3.2:3b (2.0 GB) answered 2/12 - and lost structured extraction 4/9 against 9/9.
Placed alphabetically in Testing, Evaluation and Observability, matching the existing star-badge format. Happy to move it or reword if you'd prefer it elsewhere.