Prove the lift · AI model performance evaluation
gemini-1.5-pro vs gpt-4o-mini
+18-point lift for gemini-1.5-pro on insurance-extraction-eval
Test summary
This benchmark evaluates gemini-1.5-pro against gpt-4o-mini on the insurance-extraction-eval v2 dataset (5 rows), scored by an LLM-as-judge (llama-3.1-8b). gemini-1.5-pro reached a 69% pass rate versus 51% for gpt-4o-mini — a +18-point lift.
- Base model
- gpt-4o-mini
- via openai · 51% pass rate
- Refined model
- gemini-1.5-pro
- via google · 69% pass rate · 0.7s p95
Base pass rate
51%
Refined pass rate
69%
Refined p95
0.7s
Est. tokens saved
1.3K
+0pts
refined beats base on 5 rows
0%
Base model0%
Invoked-refined▲1 improved▼0 regressed●4 unchanged
Per-row results
1Business-interruption period-of-restoration calculation100% → 100%0
2Aggregate deductible across a loss sequence under a per-occurrence + annual cap0% → 100%+100
3Replacement cost vs ACV on a depreciated roof100% → 100%0
4Coinsurance 80% clause recovery on an under-insured building100% → 100%0
5Suit-limitation clause: is a late-filed claim contractually barred?100% → 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