Finance Agent v2.1 - General Qualitative Analysis
Professional Work · 2026-10-19
Partial-credit score on the General Qualitative Analysis category of Finance Agent v2.1.
Top models (higher is better)
| Model | Score |
|---|---|
| Gemini 4 Argon | 79.2 |
| Muse Wonder Full | 76.1 |
| Muse Spark Open | 75.3 |
| MiMo-V2.6-Pro | 74.2 |
| MiniMax M3.1 Flash | 73.9 |
| GLM-5.3 | 73.6 |
| DeepSeek V4.1 Flash | 73.0 |
| Mistral Large 4 Preview | 71.7 |
| GLM-5.3 Flash | 71.4 |
| Gemini 3.8 Flash | 71.0 |
| DeepSeek V4.1 Flash 1017 | 71.0 |
| Grok 4.7 | 70.1 |
| DeepSeek V4 Pro 0813 | 69.1 |
| MiniMax M3 | 69.1 |
| Inkling | 68.6 |
| MiMo-V2.6-Flash | 68.5 |
| Beam 1.0 501B A23B | 68.2 |
| Muse Wonder | 67.5 |
| Kimi K3 | 67.0 |
| Mercury 3 | 66.9 |
| GPT-6.1 Sol | 64.5 |
| Step 5 Preview | 64.0 |
| Claude Haiku 5.5 | 63.6 |
| Hy4 Preview | 62.9 |
| GPT-6 Astra | 61.9 |
| Inkling-Small | 57.7 |
| GPT-6 Luna | 56.8 |