BizBench - FinKnow: leaderboard
Metric: Accuracy (%; 877 multiple-choice business and finance knowledge questions from CFA practice exams and three MMLU subjects, zero-shot, highest answer log-probability). Source: arxiv.org. Saturation forecast: Estimated already saturated. 16 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4 | 80.1 |
| 2 | GPT-3.5 | 60.3 |
| 3 | Mixtral 8x7B | 47.1 |
| 4 | Llama 2 70B | 44.9 |
| 5 | Mistral 7B | 44.4 |
| 6 | falcon-40B | 43.9 |
| 7 | mpt-30B | 42.4 |
| 8 | Llama 2 13B | 42.1 |
| 9 | Llama 2 7B | 41.7 |
| 10 | falcon-7B | 40.9 |
| 11 | mpt-7B | 39.7 |
| 12 | starcoder | 37.9 |
Interactive version: theaggregate.ai/benchmark?slug=bizbench-finknow · How It Works · Data refreshed daily, snapshot 2026-09-26.