feat: add the quality, latency, and cost evaluation harness - #16
Merged
Merged
Conversation
Grades committed datasets with task-appropriate scoring and structured output validity, reports quality with latency percentiles, throughput, and GPU seconds per successful request, and rejects any variant that buys latency with quality. Adds reproducible k6 steady and burst workloads.
execute() drives each committed EvaluationCase through a caller-supplied transport and measures latency the same way for every harness run, so build_report() can be fed real outcomes instead of hand-built ones. This is the piece section 16 needed to actually exercise a dataset against a model instead of only scoring pre-collected outcomes. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
github-actions
Bot
added
documentation
5 tasks
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
execute(), the transport runner that drives committedEvaluationCases through a caller-supplied callable and times each one, sobuild_report()can be fed real outcomes instead of only hand-built ones.benchmarks/workloads.Test plan
ruff format --check .ruff check .mypypytest tests/unit tests/integration --cov=llm_router --cov-report=term-missing(97% coverage, 90% required)