FinChain: leaderboard
Financial chain-of-thought benchmark reporting ChainEval and lexical/semantic similarity metrics for general and finance-specific models.
Metric: ChainEval (imported). Source: mbzuai-nlp.github.io. Source: benchmarklist.com. Saturation forecast: Around April 2027. 25 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Pro | 58.65 | #152 |
| 2 | Claude Sonnet 4.5 | 58.22 | #142 |
| 3 | Claude Sonnet 4 | 58.18 | #199 |
| 4 | Gemini 2.5 Flash | 58.01 | #245 |
| 5 | Claude 3.7 Sonnet | 57.89 | #249 |
| 6 | GPT-5 Mini | 57.38 | #186 |
| 7 | GPT-4.1 Mini | 57.24 | #355 |
| 8 | GPT-5 | 57.07 | #98 |
| 9 | GPT-4.1 | 56.92 | #248 |
| 10 | DeepSeek V3.1 | 56.76 | #264 |
| 11 | Grok 4 Fast | 56.73 | #254 |
| 12 | DeepSeek V3.2 | 56.71 | #205 |
| 13 | DeepSeek R1 | 53.75 | #251 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=finchain · How It Works · Data refreshed daily, snapshot 2026-10-20.