FinRiskAtlas - Domain Knowledge - Trade: leaderboard
Macro-averaged accuracy on the Trade 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 | Gemini 3 Flash (Preview) | 83.8 | #90 |
| 2 | Kimi K3 | 83.05 | #35 |
| 3 | GPT-5.6 Sol | 83 | #22 |
| 4 | Qwen 3.7 Max | 82.49 | #85 |
| 5 | Claude Opus 4.7 | 82.15 | #51 |
| 6 | Kimi K2.5 | 81.95 | #152 |
| 7 | GPT-5.6 Terra | 81.58 | #40 |
| 8 | GPT-5.4 | 81.17 | #83 |
| 9 | Seed 2.0 Lite | 80.36 | #161 |
| 10 | GPT-5.4 Mini | 79.48 | #228 |
| 11 | GLM-5.2 | 79.18 | #80 |
| 12 | Qwen 3 235B A22B | 79.02 | #323 |
| 13 | Qwen3-Next-80B | 78.97 | #321 |
| 14 | Ling-2.6-1T | 78.96 | #307 |
| 15 | GLM-5 | 78.14 | #149 |
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-trade · How It Works · Data refreshed daily, snapshot 2026-11-05.