Prove the lift · AI model performance evaluation

gemini-2.5-pro vs claude-opus-4-8

+18-point lift for gemini-2.5-pro on medical-coding-extract

Test summary

This benchmark evaluates gemini-2.5-pro against claude-opus-4-8 on the medical-coding-extract v2 dataset (5 rows), scored by an LLM-as-judge (gpt-4o-mini). gemini-2.5-pro reached a 62% pass rate versus 44% for claude-opus-4-8 — a +18-point lift.

Base model
claude-opus-4-8
via anthropic · 44% pass rate
Refined model
gemini-2.5-pro
via google · 62% pass rate · 1.9s p95
Base pass rate
44%
Refined pass rate
62%
Refined p95
1.9s
Est. tokens saved
8.0K
+0pts
refined beats base on 5 rows
0%
Base model
0%
Invoked-refined
0 improved0 regressed5 unchanged

Per-row results

1Pull the procedure code from an operative report100%100%0
2Extract medication + dosage from a clinical note100%100%0
3Determine the correct E/M level from a visit note100%100%0
4Identify a contraindication from a problem list0%0%0
5Map a discharge summary to the primary ICD-10 code0%0%0
Prove the lift on your own data.

Run any two models over your tasks, judged automatically, and see the improvement — in seconds.

Try Invoked
Generated with Invoked · Browse all benchmarks