Prove the lift · AI model performance evaluation

gpt-4-turbo vs mistral-large

+21-point lift for gpt-4-turbo on support-ticket-triage

Test summary

This benchmark evaluates gpt-4-turbo against mistral-large on the support-ticket-triage v2 dataset (5 rows), scored by an LLM-as-judge (gpt-4o-mini). gpt-4-turbo reached a 74% pass rate versus 53% for mistral-large — a +21-point lift.

Base model
mistral-large
via mistral · 53% pass rate
Refined model
gpt-4-turbo
via openai · 74% pass rate · 0.7s p95
Base pass rate
53%
Refined pass rate
74%
Refined p95
0.7s
Est. tokens saved
9.8K
+0pts
refined beats base on 5 rows
0%
Base model
0%
Invoked-refined
5 improved0 regressed0 unchanged

Per-row results

1Route a multi-part ticket to the correct queue0%100%+100
2Classify: billing dispute vs refund request vs cancellation0%100%+100
3Priority-tag an outage report vs a feature question0%100%+100
4Detect churn-risk sentiment in a renewal thread0%100%+100
5Extract the affected order id from a free-text complaint0%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 Invoked
Generated with Invoked · Browse all benchmarks