Add benchmark results for moonshotai/kimi-linear-48b-a3b-instruct

[github-actions]

This PR adds benchmark results for the moonshotai/kimi-linear-48b-a3b-instruct model.

The following files have been updated:

• src/benchmark/results.json - Raw benchmark results
• src/benchmark/validation-results.json - Validation results against human baseline

This PR was automatically generated by the benchmark workflow.

Note: If you don't want to merge this PR, close it and the model will be added to the untested list to prevent re-processing.

@alrocar

---

Note

Adds raw benchmark runs and validation outcomes for moonshotai/kimi-linear-48b-a3b-instruct, and registers the model in the benchmark config.

• Benchmark config:
  • Register moonshotai/kimi-linear-48b-a3b-instruct under moonshotai in src/benchmark-config.json.
• Results:
  • Append comprehensive raw run logs for the model across many queries/pipes in src/benchmark/results.json (SQLs, outputs, errors, metrics, attempts).
• Validation:
  • Update src/benchmark/validation-results.json with per-task comparisons for the new model (match/non-match statuses, distances, SQL used) and aggregate metrics.

^{Written by Cursor Bugbot for commit 2f7677a. This will update automatically on new commits. Configure here.}

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Preview	Comments	Updated (UTC)
llm-benchmark	Ready	Preview	Comment	Nov 8, 2025 6:09am