Prove the lift · AI model performance evaluation

grok-4 vs claude-opus-4-8

+17-point lift for grok-4 on ecommerce-catalog-norm

Test summary

This benchmark evaluates grok-4 against claude-opus-4-8 on the ecommerce-catalog-norm v1 dataset (5 rows), scored by an LLM-as-judge (claude-haiku-4-5). grok-4 reached a 51% pass rate versus 34% for claude-opus-4-8 — a +17-point lift.

Base model
claude-opus-4-8
via anthropic · 34% pass rate
Refined model
grok-4
via xai · 51% pass rate · 0.8s p95
Base pass rate
34%
Refined pass rate
51%
Refined p95
0.8s
Est. tokens saved
2.4K
+0pts
refined beats base on 5 rows
0%
Base model
0%
Invoked-refined
0 improved1 regressed3 unchanged

Per-row results

1Extract material + color from a product title100%100%0
2Deduplicate two near-identical SKUs100%0%-100
3Normalize a free-text size into a standard enum0%0%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