Flame Corporate Finance - Evaluate: leaderboard

Metric: Accuracy (%; the paper's Bloom-level weighted accuracy on the Evaluate problems of the 685-problem textbook-grounded Corporate Finance multiple-choice set, zero-shot, temperature 0). Source: arxiv.org. Saturation forecast: Estimated already saturated. 12 models tracked.

Top models

#ModelScore
1Claude Opus 4.695
2Gemini 3.1 Pro (Preview)93
3GPT-5.482
4Gemini 2.5 Flash Lite79
5GPT-4.1 Mini74
6Qwen 3 32B71
7Claude Haiku 4.565
8Gemma 3 27B (IT)58
9DeepSeek R1 Distill Qwen 32B50
10Gemma 3 12B (IT)46
11Qwen 3 8B43
12DeepSeek R1 Distill Qwen 14B43

Interactive version: theaggregate.ai/benchmark?slug=flame-corporate-finance-evaluate · How It Works · Data refreshed daily, snapshot 2026-09-26.