Finance Agent v2.1 - Disclosure Analysis
Professional Work · 2026-10-19
Partial-credit score on the Disclosure Analysis category of Finance Agent v2.1.
Top models (higher is better)
| Model | Score |
|---|---|
| Gemini 4 Argon | 83.3 |
| Muse Wonder Full | 80.6 |
| Muse Wonder | 78.2 |
| Hy4 Preview | 77.6 |
| GLM-5.3 | 77.0 |
| DeepSeek V4.1 Flash | 76.9 |
| Muse Spark Open | 76.8 |
| Gemini 3.8 Flash | 76.6 |
| MiMo-V2.6-Flash | 75.4 |
| MiMo-V2.6-Pro | 75.3 |
| Mistral Large 4 Preview | 73.5 |
| Claude Haiku 5.5 | 71.8 |
| Kimi K3 | 71.6 |
| Grok 4.7 | 71.2 |
| GLM-5.3 Flash | 70.5 |
| MiniMax M3.1 Flash | 70.5 |
| DeepSeek V4.1 Flash 1017 | 70.3 |
| MiniMax M3 | 70.1 |
| Step 5 Preview | 68.4 |
| GPT-6.1 Sol | 68.3 |
| Beam 1.0 501B A23B | 66.1 |
| DeepSeek V4 Pro 0813 | 65.1 |
| GPT-6 Astra | 64.7 |
| Inkling | 64.3 |
| Mercury 3 | 63.3 |
| GPT-6 Luna | 57.6 |
| Inkling-Small | 54.6 |