Prove the lift · AI model performance evaluation
glm-5.2 vs gpt-oss-120b
+21-point lift for glm-5.2 on ecommerce-catalog-norm
Test summary
This benchmark evaluates glm-5.2 against gpt-oss-120b on the ecommerce-catalog-norm v1 dataset (5 rows), scored by an LLM-as-judge (gpt-4o-mini). glm-5.2 reached a 71% pass rate versus 50% for gpt-oss-120b — a +21-point lift.
- Base model
- gpt-oss-120b
- via openai · 50% pass rate
- Refined model
- glm-5.2
- via z-ai · 71% pass rate · 2.5s p95
Base pass rate
50%
Refined pass rate
71%
Refined p95
2.5s
Est. tokens saved
7.2K
+0pts
refined beats base on 5 rows
0%
Base model0%
Invoked-refined▲3 improved▼0 regressed●1 unchanged
Per-row results
1Classify a listing into the right category node0% → 100%+100
2Extract material + color from a product title100% → 100%0
3Deduplicate two near-identical SKUs0% → 100%+100
4Normalize a free-text size into a standard enum0% → 100%+100
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