Prove the lift · AI model performance evaluation

claude-sonnet-4-6 vs gpt-4o-mini

-6-point lift for claude-sonnet-4-6 on contract-clause-qa

Test summary

This benchmark evaluates claude-sonnet-4-6 against gpt-4o-mini on the contract-clause-qa v1 dataset (5 rows), scored by an LLM-as-judge (gpt-4o-mini). claude-sonnet-4-6 reached a 36% pass rate versus 42% for gpt-4o-mini — a -6-point lift.

Base model
gpt-4o-mini
via openai · 42% pass rate
Refined model
claude-sonnet-4-6
via anthropic · 36% pass rate · 2.6s p95
Base pass rate
42%
Refined pass rate
36%
Refined p95
2.6s
Est. tokens saved
—
−0pts
no lift — the base model held its ground
0%
Base model
0%
Invoked-refined
▲2 improved▼0 regressed●3 unchanged

Per-row results

1Determine indemnification scope and carve-outs0% → 0%0
2Extract the auto-renewal notice window in days0% → 100%+100
3Find the liability cap and its exceptions0% → 100%+100
4Identify the governing-law clause and jurisdiction0% → 0%0
5Locate the termination-for-convenience terms100% → 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