FinRiskAtlas - Domain Knowledge - Finance & Law: leaderboard
Macro-averaged accuracy on the Finance & Law 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 | GPT-5.4 | 85.34 | #83 |
| 2 | Qwen 3.7 Max | 83.65 | #85 |
| 3 | Claude Opus 4.7 | 82.75 | #51 |
| 4 | Qwen3-Next-80B | 82.08 | #321 |
| 5 | Nemotron 3 Ultra | 81.85 | #245 |
| 6 | Qwen 3 235B A22B | 81.26 | #323 |
| 7 | Kimi K3 | 80.95 | #35 |
| 8 | Qwen 3 32B | 80.66 | #449 |
| 9 | Gemini 3 Flash (Preview) | 80.53 | #90 |
| 10 | Ling-2.6-1T | 80.49 | #307 |
| 11 | GPT-5.6 Terra | 80.24 | #40 |
| 12 | GPT-5.4 Mini | 80.16 | #228 |
| 13 | Seed 2.0 Mini | 80.04 | #279 |
| 14 | DeepSeek V4 Pro | 79.88 | #112 |
| 15 | DeepSeek V3 (0324) | 79.48 | #349 |
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-finance-law · How It Works · Data refreshed daily, snapshot 2026-11-05.