FINESSE-Bench: leaderboard
Large language models (LLMs) are increasingly being applied to financial analysis, reporting, investment decision support, risk management, compliance, and professional training. However, robust evaluation of their domain competence in finance remains incomplete. Widely used open benchmarks such as FinQA, ConvFinQA, and TAT-QA have played an important role in advancing financial question answering
Metric: exam-like (self-reported). Source: benchmarklist.com. Saturation forecast: Around May 2027. 31 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3.5 Plus (02-15) | 87.76 | #218 |
| 2 | Claude Sonnet 4.6 | 87.63 | #90 |
| 3 | Kimi K2.5 | 87.4 | #144 |
| 4 | GLM-5 | 87.27 | #141 |
| 5 | GLM-4.7 | 86.39 | #192 |
| 6 | GPT-5.2 | 85.95 | #110 |
| 7 | Qwen 3.5 397B A17B | 84.43 | #146 |
| 8 | Qwen 3.5 27B | 84.35 | #187 |
| 9 | Qwen 3.5 Flash (02-23) | 84.08 | #185 |
| 10 | Qwen 3.5 35B A3B | 83.9 | #256 |
| 11 | Qwen 3 235B A22B 2507 (Thinking) | 83.81 | #260 (Qwen 3 235B A22B 2507) |
| 12 | DeepSeek R1 0528 | 82.66 | #222 |
| 13 | Claude 3.7 Sonnet | 82.61 | #249 |
| 14 | MiniMax-M2.5 | 80.05 | #303 |
| 15 | Qwen 3 32B | 79.38 | #443 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=finesse-bench · How It Works · Data refreshed daily, snapshot 2026-10-20.