Finance Agent v2 - Financial Modeling
Professional Work · 2026-05-19
The Financial Modeling 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 |
|---|---|
| Muse Spark 1.1 | 34.1 |
| Gemini 3.5 Flash | 29.4 |
| Claude Opus 5 | 26.6 |
| Gemini 3.6 Flash | 25.1 |
| Sonnet 5 | 24.2 |
| Opus 4.8 | 23.3 |
| Kimi K3 | 23.0 |
| GPT-5.5 | 22.9 |
| GPT-5.6 Luna | 21.6 |
| GPT-5.6 Terra | 21.4 |
| GPT-5.6 Sol | 20.7 |
| Opus 4.7 | 20.1 |
| GPT-5.4 Mini | 19.6 |
| Gemini 3.5 Flash-Lite | 19.3 |
| Claude Fable 5 | 17.6 |
| Sonnet 4.6 | 17.6 |
| Qwen3.7-Max | 17.0 |
| Qwen3.6 Plus (2026-04-02) | 16.7 |
| Inkling | 16.7 |
| Grok 4.5 | 15.6 |
| Gemini 3.1 Pro Preview | 15.1 |
| Gemini 3 Flash Preview | 14.4 |
| Kimi K2.6 | 14.2 |
| MiniMax M3 | 14.0 |
| GPT-5.4 Nano | 13.5 |
| GLM-5.1 | 13.5 |
| GLM-5.2 | 12.9 |
| DeepSeek-V4-Pro | 10.5 |
| Qwen3.7-Plus | 10.1 |
| MiMo-V2.5-Pro | 9.9 |
| Nemotron 3 Ultra 550B A55B | 7.1 |
| Grok 4.3 | 5.9 |
| MiMo-V2.5 | 5.3 |
| Haiku 4.5 | 5.1 |
| Mistral Medium 3.5 | 5.1 |
| Gemini 3.1 Flash-Lite Preview | 4.1 |
| MiniMax M2.7 | 4.1 |
| Laguna M.1 | 2.6 |
| Laguna XS.2 | 2.6 |
| Grok 4.20 | 2.4 |