Prove the lift · AI model performance evaluation
gpt-4o vs o3-mini
+2-point lift for gpt-4o on ecommerce-catalog-norm
Test summary
This benchmark evaluates gpt-4o against o3-mini on the ecommerce-catalog-norm v2 dataset (5 rows), scored by an LLM-as-judge (gpt-4o-mini, run locally). gpt-4o reached a 49% pass rate versus 47% for o3-mini — a +2-point lift.
- Base model
- o3-mini
- via openai · 47% pass rate
- Refined model
- gpt-4o
- via openai · 49% pass rate · 1.3s p95
Base pass rate
47%
Refined pass rate
49%
Refined p95
1.3s
Est. tokens saved
5.8K
+0pts
refined beats base on 5 rows
0%
Base model0%
Invoked-refined▲1 improved▼0 regressed●3 unchanged
Per-row results
1Classify a listing into the right category node0% → 100%+100
2Deduplicate two near-identical SKUs100% → 100%0
3Extract material + color from a product title100% → 100%0
4Normalize a free-text size into a standard enum100% → 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