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 |
|---|---|
| Claude Opus 5 | 74.2 |
| Gemini 3.5 Flash | 74.0 |
| Gemini 3.6 Flash | 69.8 |
| Claude Fable 5 | 68.9 |
| Muse Spark 1.1 | 67.2 |
| Opus 4.8 | 67.2 |
| Kimi K3 | 66.6 |
| GPT-5.6 Luna | 64.9 |
| GPT-5.6 Sol | 64.2 |
| Sonnet 5 | 64.1 |
| Opus 4.7 | 62.7 |
| GLM-5.2 | 61.5 |
| GPT-5.5 | 60.3 |
| GPT-5.6 Terra | 60.1 |
| Qwen3.7-Max | 59.5 |
| MiniMax M3 | 59.0 |
| Gemini 3 Flash Preview | 59.0 |
| Sonnet 4.6 | 58.5 |
| Grok 4.5 | 57.8 |
| Gemini 3.5 Flash-Lite | 57.0 |
| GPT-5.4 Mini | 56.9 |
| Inkling | 56.6 |
| Kimi K2.6 | 53.1 |
| Gemini 3.1 Pro Preview | 52.8 |
| GLM-5.1 | 51.9 |
| DeepSeek-V4-Pro | 50.8 |
| Grok 4.3 | 50.2 |
| MiMo-V2.5-Pro | 49.0 |
| MiMo-V2.5 | 48.6 |
| Qwen3.7-Plus | 48.5 |
| Qwen3.6 Plus (2026-04-02) | 47.3 |
| GPT-5.4 Nano | 46.5 |
| Nemotron 3 Ultra 550B A55B | 41.3 |
| Haiku 4.5 | 40.7 |
| Mistral Medium 3.5 | 37.0 |
| Laguna M.1 | 35.6 |
| Gemini 3.1 Flash-Lite Preview | 34.2 |
| MiniMax M2.7 | 33.4 |
| Grok 4.20 | 33.3 |
| Laguna XS.2 | 20.4 |