Finance Agent v2.1 - Adjustments
Professional Work · 2026-10-19
Partial-credit score on the Adjustments category of Finance Agent v2.1.
Top models (higher is better)
| Model | Score |
|---|---|
| Muse Wonder Full | 74.7 |
| Gemini 4 Argon | 70.8 |
| Claude Haiku 5.5 | 67.7 |
| GLM-5.3 | 67.5 |
| DeepSeek V4.1 Flash 1017 | 66.8 |
| Hy4 Preview | 66.7 |
| DeepSeek V4.1 Flash | 66.6 |
| Muse Wonder | 64.7 |
| Muse Spark Open | 63.0 |
| Gemini 3.8 Flash | 61.9 |
| Kimi K3 | 61.8 |
| MiniMax M3.1 Flash | 59.7 |
| Mistral Large 4 Preview | 59.6 |
| MiniMax M3 | 59.5 |
| MiMo-V2.6-Pro | 59.1 |
| Grok 4.7 | 58.5 |
| Step 5 Preview | 58.5 |
| GPT-6 Astra | 58.1 |
| GLM-5.3 Flash | 57.6 |
| MiMo-V2.6-Flash | 57.3 |
| GPT-6 Luna | 56.3 |
| GPT-6.1 Sol | 54.7 |
| Mercury 3 | 54.1 |
| DeepSeek V4 Pro 0813 | 51.4 |
| Beam 1.0 501B A23B | 44.9 |
| Inkling | 42.0 |
| Inkling-Small | 39.4 |