Finance Agent v2 - Comparables
Professional Work · 2026-05-19
The Comparables 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 3.8 Flash | 52.0 |
| Gemini 4 Argon | 51.0 |
| Gemini 3.7 Flash | 50.3 |
| Muse Spark 1.2 | 50.2 |
| MiMo-V2.6-Flash | 49.0 |
| Gemini 3.5 Flash | 48.8 |
| Mistral Large 4 Preview | 48.6 |
| GPT-5.6 Luna | 48.3 |
| Muse Wonder | 47.9 |
| Gemini 3.6 Flash | 47.8 |
| Kimi K3 | 47.8 |
| Muse Spark Open | 47.6 |
| DeepSeek V4.1 Flash 1017 | 47.5 |
| Muse Spark 1.3 | 47.2 |
| Muse Spark 1.1 | 47.2 |
| GPT-5.6 Terra | 47.0 |
| Muse Wonder Full | 46.8 |
| Qwen3.8 Max (0902) | 46.6 |
| Step 5 Preview | 46.4 |
| DeepSeek V4 Pro 0813 | 46.2 |
| DeepSeek V4.1 Flash | 46.1 |
| MiMo-V2.6-Pro | 46.0 |
| Claude Fable 5.1 | 46.0 |
| GLM-5.3 Flash | 45.9 |
| Sonnet 5 | 45.8 |
| GLM-5.3 | 45.7 |
| Opus 4.7 | 45.6 |
| Claude Haiku 5.5 | 45.3 |
| GPT-5.6 Sol | 45.3 |
| Gemini 3.5 Flash-Lite | 45.2 |
| Opus 4.8 | 44.9 |
| GLM-5.2 | 44.8 |
| Hy4 Preview | 44.4 |
| Qwen3.8-Max | 44.3 |
| Qwen3.6 Plus (2026-04-02) | 43.7 |
| GPT-5.5 | 43.7 |
| Inkling | 42.8 |
| GPT-6.1 Sol | 42.6 |
| Qwen3.7-Max (source unspecified) | 42.3 |
| Grok 4.7 | 42.2 |