Finance Agent v2 - General Quantitative Analysis
Professional Work · 2026-05-19
The General Quantitative 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 3.6 Flash | 81.8 |
| Claude Fable 5 | 80.8 |
| Claude Opus 5 | 78.5 |
| Gemini 3.5 Flash | 77.6 |
| Muse Spark 1.1 | 76.8 |
| GPT-5.6 Sol | 75.6 |
| GPT-5.6 Luna | 75.2 |
| Kimi K3 | 74.4 |
| Sonnet 5 | 74.3 |
| GPT-5.5 | 73.7 |
| Sonnet 4.6 | 72.0 |
| GPT-5.6 Terra | 71.9 |
| Opus 4.8 | 71.6 |
| Opus 4.7 | 70.1 |
| Qwen3.7-Max | 67.1 |
| Gemini 3.1 Pro Preview | 67.0 |
| MiniMax M3 | 66.5 |
| Gemini 3.5 Flash-Lite | 66.1 |
| DeepSeek-V4-Pro | 65.8 |
| GLM-5.2 | 65.5 |
| Kimi K2.6 | 65.1 |
| Gemini 3 Flash Preview | 64.5 |
| Inkling | 64.2 |
| GPT-5.4 Mini | 64.0 |
| Grok 4.5 | 63.6 |
| GLM-5.1 | 61.9 |
| MiMo-V2.5-Pro | 61.0 |
| GPT-5.4 Nano | 58.3 |
| Qwen3.6 Plus (2026-04-02) | 58.2 |
| Nemotron 3 Ultra 550B A55B | 55.8 |
| MiMo-V2.5 | 55.7 |
| Qwen3.7-Plus | 52.2 |
| Grok 4.3 | 51.8 |
| Gemini 3.1 Flash-Lite Preview | 49.9 |
| Haiku 4.5 | 46.0 |
| Mistral Medium 3.5 | 44.8 |
| MiniMax M2.7 | 43.0 |
| Laguna M.1 | 39.3 |
| Grok 4.20 | 37.2 |
| Laguna XS.2 | 22.7 |