Prove the lift · AI model performance evaluation

llama-3.3-70b vs gpt-4o

+11-point lift for llama-3.3-70b on contract-clause-qa

Test summary

This benchmark evaluates llama-3.3-70b against gpt-4o on the contract-clause-qa v1 dataset (5 rows), scored by an LLM-as-judge (claude-haiku-4-5, run locally). llama-3.3-70b reached a 62% pass rate versus 51% for gpt-4o — a +11-point lift.

Base model
gpt-4o
via openai · 51% pass rate
Refined model
llama-3.3-70b
via meta · 62% pass rate · 1.4s p95
Base pass rate
51%
Refined pass rate
62%
Refined p95
1.4s
Est. tokens saved
3.6K
+0pts
refined beats base on 5 rows
0%
Base model
0%
Invoked-refined
0 improved0 regressed5 unchanged

Per-row results

1Identify the governing-law clause and jurisdiction100%100%0
2Extract the auto-renewal notice window in days100%100%0
3Determine indemnification scope and carve-outs100%100%0
4Find the liability cap and its exceptions100%100%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