Add benchmark results for mistralai/ministral-14b-2512

[github-actions]

This PR adds benchmark results for the mistralai/ministral-14b-2512 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 benchmark results and validation comparisons for mistralai/ministral-14b-2512, and includes the model in the benchmark config.

• Benchmark Config:
  • Add mistralai model ministral-14b-2512 to src/benchmark-config.json.
• Benchmarks:
  • Append extensive raw runs for mistralai/ministral-14b-2512 in src/benchmark/results.json across many SQL tasks (queries, attempts, timings, tokens, errors).
• Validation:
  • Update src/benchmark/validation-results.json with per-task comparisons for mistralai/ministral-14b-2512, including SQL used, row counts, match metrics, and aggregate stats (model shows no matches).

^{Written by Cursor Bugbot for commit eb7a7df. 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	Dec 5, 2025 0:20am