FinVerBench: leaderboard
Metric: Accuracy (self-reported). Source: benchmarklist.com. 15 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 4 | 100 |
| 2 | Qwen 3 235B A22B | 65.7 |
| 3 | Gemma 3 27B | 61.9 |
| 4 | GPT-4.1 | 61 |
| 5 | DeepSeek R1 | 60 |
| 6 | Claude Sonnet 4.6 | 59 |
| 7 | GPT-5.2 | 59 |
| 8 | GPT-5.5 | 59 |
| 9 | GPT-5.4 | 59 |
| 10 | Llama 4 Maverick | 59 |
| 11 | Claude Opus 4.6 | 59 |
| 12 | DeepSeek V3.2 | 59 |
| 13 | MiniMax-M2.5 | 59 |
| 14 | Llama 4 Scout | 58.1 |
Interactive version: theaggregate.ai/benchmark?slug=finverbench · How It Works · Data refreshed daily, snapshot 2026-09-05.