Prove the lift · AI model performance evaluation
nova-pro vs claude-3-7-sonnet
+18-point lift for nova-pro on medical-coding-extract
Test summary
This benchmark evaluates nova-pro against claude-3-7-sonnet on the medical-coding-extract v1 dataset (5 rows), scored by an LLM-as-judge (gpt-4o-mini). nova-pro reached a 57% pass rate versus 39% for claude-3-7-sonnet — a +18-point lift.
- Base model
- claude-3-7-sonnet
- via anthropic · 39% pass rate
- Refined model
- nova-pro
- via aws · 57% pass rate · 2.0s p95
Base pass rate
39%
Refined pass rate
57%
Refined p95
2.0s
Est. tokens saved
8.0K
+0pts
refined beats base on 5 rows
0%
Base model0%
Invoked-refined▲3 improved▼0 regressed●2 unchanged
Per-row results
1Map a discharge summary to the primary ICD-10 code0% → 100%+100
2Extract medication + dosage from a clinical note100% → 100%0
3Identify a contraindication from a problem list0% → 100%+100
4Pull the procedure code from an operative report0% → 100%+100
5Determine the correct E/M level from a visit note100% → 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