Prove the lift · AI model performance evaluation
claude-opus-4-8 vs mixtral-8x7b
+1-point lift for claude-opus-4-8 on medical-coding-extract
Test summary
This benchmark evaluates claude-opus-4-8 against mixtral-8x7b on the medical-coding-extract v1 dataset (5 rows), scored by an LLM-as-judge (gpt-4o-mini). claude-opus-4-8 reached a 30% pass rate versus 29% for mixtral-8x7b — a +1-point lift.
- Base model
- mixtral-8x7b
- via mistral · 29% pass rate
- Refined model
- claude-opus-4-8
- via anthropic · 30% pass rate · 2.2s p95
Base pass rate
29%
Refined pass rate
30%
Refined p95
2.2s
Est. tokens saved
5.7K
+0pts
refined beats base on 5 rows
0%
Base model0%
Invoked-refined▲1 improved▼1 regressed●3 unchanged
Per-row results
1Determine the correct E/M level from a visit note100% → 100%0
2Extract medication + dosage from a clinical note100% → 0%-100
3Map a discharge summary to the primary ICD-10 code100% → 100%0
4Pull the procedure code from an operative report0% → 100%+100
5Identify a contraindication from a problem list100% → 100%0
Prove the lift on your own data.
Run any two models over your tasks, judged automatically, and see the improvement — in seconds.
Try InvokedGenerated with Invoked · Browse all benchmarks