Finance Agent v2 - Disclosure Analysis
Professional Work · 2026-05-19
The Disclosure 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 | 74.5 |
| Muse Wonder Full | 71.6 |
| GLM-5.3 Flash | 71.5 |
| MiMo-V2.6-Flash | 70.9 |
| Claude Fable 5.1 | 70.3 |
| Muse Spark 1.3 | 70.1 |
| GLM-5.3 | 69.6 |
| Muse Spark 1.2 | 69.5 |
| Hy4 Preview | 68.8 |
| Qwen3.8 Max (0902) | 68.6 |
| Gemini 3.8 Flash | 68.5 |
| Muse Spark Open | 67.8 |
| Gemini 3.7 Flash | 67.7 |
| Muse Wonder | 67.3 |
| GPT-5.6 Terra | 66.9 |
| MiMo-V2.6-Pro | 66.7 |
| DeepSeek V4.1 Flash 1017 | 66.3 |
| Sonnet 4.6 | 65.8 |
| Opus 4.8 | 65.7 |
| Kimi K3 | 65.1 |
| Grok 4.7 | 64.5 |
| Grok 4.6 | 64.4 |
| Claude Haiku 5.5 | 64.3 |
| Gemini 3.5 Flash | 63.9 |
| Mistral Large 4 Preview | 63.9 |
| Muse Spark 1.1 | 63.4 |
| GPT-5.6 Sol | 63.3 |
| DeepSeek-V4-Flash-0731 | 63.3 |
| GPT-5.6 Luna | 62.8 |
| Sonnet 5 | 62.1 |
| Gemini 3.6 Flash | 62.0 |
| GPT-6 Astra | 61.7 |
| GLM-5.2 | 61.6 |
| Opus 4.7 | 61.3 |
| DeepSeek V4.1 Flash | 61.3 |
| Step 5 Preview | 61.3 |
| Ember-1 | 61.1 |
| Qwen3.8-Max | 60.5 |
| Qwen3.8 27B | 60.3 |
| DeepSeek V4 Pro 0813 | 59.9 |