FinRiskAtlas - Domain Knowledge - Commerce & Payment: leaderboard
Macro-averaged accuracy on the Commerce & Payment group of FinRiskAtlas domain-knowledge task families, a subset of the 42-family knowledge layer used for coarse screening of financial-review models (FinRiskAtlas, arXiv 2608.25325v1, Appendix Table 8). Percent; higher is better; zero-shot direct answers from 33 archived configurations.
Metric: Macro-Averaged Accuracy (%). Source: arxiv.org. Saturation forecast: Around February 2027. 33 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.7 | 86.81 | #51 |
| 2 | Gemini 3 Flash (Preview) | 86.41 | #90 |
| 3 | Qwen 3.7 Max | 86.01 | #85 |
| 4 | GPT-5.4 | 84.34 | #83 |
| 5 | GPT-5.4 Mini | 84.34 | #228 |
| 6 | Kimi K3 | 82.8 | #35 |
| 7 | GPT-5.6 Terra | 82.73 | #40 |
| 8 | GPT-5.6 Sol | 82.42 | #22 |
| 9 | Qwen 3 235B A22B | 82.08 | #323 |
| 10 | DeepSeek V4 Pro | 81.55 | #112 |
| 11 | Seed 2.0 Lite | 81.51 | #161 |
| 12 | Gemini 2.5 Flash | 80.51 | #255 |
| 13 | Kimi K2.5 | 80.25 | #152 |
| 14 | GLM-5 | 79.93 | #149 |
| 15 | Nemotron 3 Ultra | 79.73 | #245 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=finriskatlas-domain-knowledge-commerce-payment · How It Works · Data refreshed daily, snapshot 2026-11-05.