Prove the lift · AI model performance evaluation

claude-sonnet-5 vs llama-4-maverick

+8-point lift for claude-sonnet-5 on insurance-extraction-eval

Test summary

This benchmark evaluates claude-sonnet-5 against llama-4-maverick on the insurance-extraction-eval v1 dataset (7 rows), scored by an LLM-as-judge (llama-3.1-8b). claude-sonnet-5 reached a 52% pass rate versus 44% for llama-4-maverick — a +8-point lift.

Base model
llama-4-maverick
via meta · 44% pass rate
Refined model
claude-sonnet-5
via anthropic · 52% pass rate · 1.4s p95
Base pass rate
44%
Refined pass rate
52%
Refined p95
1.4s
Est. tokens saved
5.3K
+0pts
refined beats base on 7 rows
0%
Base model
0%
Invoked-refined
▲2 improved▼1 regressed●4 unchanged

Per-row results

1Suit-limitation clause: is a late-filed claim contractually barred?0% → 100%+100
2Business-interruption period-of-restoration calculation0% → 0%0
3Coinsurance 80% clause recovery on an under-insured building100% → 0%-100
4Pro-rata distribution across layered carriers on a covered loss0% → 100%+100
5Replacement cost vs ACV on a depreciated roof100% → 100%0
6Aggregate deductible across a loss sequence under a per-occurrence + annual cap100% → 100%0
7Named-peril coverage determination under a basic HO form100% → 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