Prove the lift · AI model performance evaluation

llama-4-maverick vs gpt-5.1

-12-point lift for llama-4-maverick on ecommerce-catalog-norm

Test summary

This benchmark evaluates llama-4-maverick against gpt-5.1 on the ecommerce-catalog-norm v1 dataset (5 rows), scored by an LLM-as-judge (gpt-4o-mini). llama-4-maverick reached a 37% pass rate versus 49% for gpt-5.1 — a -12-point lift.

Base model
gpt-5.1
via openai · 49% pass rate
Refined model
llama-4-maverick
via meta · 37% pass rate · 2.1s p95
Base pass rate
49%
Refined pass rate
37%
Refined p95
2.1s
Est. tokens saved
1.2K
0pts
no lift — the base model held its ground
0%
Base model
0%
Invoked-refined
1 improved0 regressed3 unchanged

Per-row results

1Normalize a free-text size into a standard enum100%100%0
2Deduplicate two near-identical SKUs0%100%+100
3Extract material + color from a product title100%100%0
4Classify a listing into the right category node100%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 Invoked
Generated with Invoked · Browse all benchmarks