Finance Agent v2.1 - Earnings Analysis
Professional Work · 2026-10-19
Partial-credit score on the Earnings Analysis category of Finance Agent v2.1.
Top models (higher is better)
| Model | Score |
|---|---|
| DeepSeek V4.1 Flash 1017 | 93.0 |
| GLM-5.3 | 92.4 |
| Muse Wonder | 90.3 |
| MiMo-V2.6-Pro | 88.5 |
| Gemini 3.8 Flash | 88.4 |
| DeepSeek V4.1 Flash | 88.3 |
| Step 5 Preview | 85.1 |
| MiMo-V2.6-Flash | 85.0 |
| Muse Wonder Full | 84.8 |
| Claude Haiku 5.5 | 84.5 |
| Grok 4.7 | 84.5 |
| GLM-5.3 Flash | 84.3 |
| GPT-6 Luna | 84.0 |
| GPT-6.1 Sol | 83.9 |
| Gemini 4 Argon | 82.6 |
| Hy4 Preview | 82.5 |
| MiniMax M3.1 Flash | 82.1 |
| Mercury 3 | 82.0 |
| Muse Spark Open | 81.3 |
| Mistral Large 4 Preview | 80.5 |
| Kimi K3 | 80.4 |
| GPT-6 Astra | 80.1 |
| MiniMax M3 | 74.4 |
| DeepSeek V4 Pro 0813 | 74.3 |
| Beam 1.0 501B A23B | 71.8 |
| Inkling-Small | 64.9 |
| Inkling | 63.8 |