Finance Agent v2 - Market Analysis
Professional Work · 2026-05-19
The Market Analysis split of Finance Agent v2. This child benchmark separates a source-reported subtask or subtrack from the parent aggregate so scores at different grains do not share one benchmark_slug.
Top models (higher is better)
| Model | Score |
|---|---|
| Gemini 4 Argon | 79.7 |
| Gemini 3.8 Flash | 78.5 |
| Gemini 3.7 Flash | 74.5 |
| Gemini 3.5 Flash | 74.0 |
| Muse Spark Open | 71.1 |
| GLM-5.3 Flash | 70.4 |
| Muse Spark 1.3 | 69.9 |
| Gemini 3.6 Flash | 69.8 |
| Qwen3.8 Max (0902) | 69.7 |
| Claude Fable 5.1 | 69.5 |
| GLM-5.3 | 67.4 |
| Muse Spark 1.1 | 67.2 |
| Opus 4.8 | 67.2 |
| Muse Spark 1.2 | 66.2 |
| GPT-6 Astra | 65.5 |
| Muse Wonder Full | 64.9 |
| GPT-5.6 Luna | 64.9 |
| GPT-6.1 Sol | 64.5 |
| GPT-5.6 Sol | 64.2 |
| Sonnet 5 | 64.1 |
| Qwen3.8-Max | 63.7 |
| MiMo-V2.6-Pro | 63.5 |
| MiMo-V2.6-Flash | 63.4 |
| DeepSeek V4.1 Flash | 63.4 |
| Claude Haiku 5.5 | 63.1 |
| Mistral Large 4 Preview | 63.0 |
| DeepSeek V4.1 Flash 1017 | 62.7 |
| Opus 4.7 | 62.7 |
| Grok 4.6 | 62.2 |
| Hy4 Preview | 62.1 |
| Muse Wonder | 62.1 |
| Kimi K3 | 62.1 |
| Grok 4.7 | 61.8 |
| GLM-5.2 | 61.5 |
| GPT-5.6 Terra | 60.6 |
| GPT-5.5 | 60.3 |
| DeepSeek V4 Pro 0813 | 60.3 |
| Qwen3.7-Max (source unspecified) | 59.5 |
| MiniMax M3 | 59.0 |
| Gemini 3 Flash Preview | 59.0 |