Prove the lift · AI model performance evaluation
nemotron-ultra-253b vs claude-3-7-sonnet
+0-point lift for nemotron-ultra-253b on contract-clause-qa
Test summary
This benchmark evaluates nemotron-ultra-253b against claude-3-7-sonnet on the contract-clause-qa v3 dataset (5 rows), scored by an LLM-as-judge (gpt-4o-mini). nemotron-ultra-253b reached a 49% pass rate versus 49% for claude-3-7-sonnet — a +0-point lift.
- Base model
- claude-3-7-sonnet
- via anthropic · 49% pass rate
- Refined model
- nemotron-ultra-253b
- via nvidia · 49% pass rate · 1.1s p95
Base pass rate
49%
Refined pass rate
49%
Refined p95
1.1s
Est. tokens saved
—
0pts
no measurable change
0%
Base model0%
Invoked-refined▲3 improved▼0 regressed●2 unchanged
Per-row results
1Determine indemnification scope and carve-outs100% → 100%0
2Find the liability cap and its exceptions0% → 100%+100
3Extract the auto-renewal notice window in days0% → 100%+100
4Locate the termination-for-convenience terms0% → 0%0
5Identify the governing-law clause and jurisdiction0% → 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