Fin-RATE: leaderboard
Financial analytics benchmark over SEC filings covering detail/reasoning QA, enterprise comparison QA, and longitudinal tracking QA.
Metric: Macro Accuracy (imported). Source: github.com. Source: benchmarklist.com. Saturation forecast: Around May 2027. 17 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4.1 | 33.24 | #248 |
| 2 | Qwen 3 235B A22B | 24.39 | #313 |
| 3 | GPT-OSS-20B | 18.69 | #518 |
| 4 | MiMo-V2-Flash | 18.05 | #372 |
| 5 | Qwen 3 30B A3B 2507 Instruct | 17.63 | #488 |
| 6 | Llama 3.3 70B Instruct | 16.76 | #534 |
| 7 | DeepSeek V3.2 | 16.32 | #205 |
| 8 | DeepSeek R1 | 15.53 | #251 |
| 9 | Qwen 3 14B | 11.25 | #537 |
| 10 | DeepSeek V3 | 9.81 | #321 |
| 11 | Qwen 3 8B | 5.48 | #674 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=fin-rate · How It Works · Data refreshed daily, snapshot 2026-10-20.