CorpFin v2 - Exact Pages
Professional Work · 2025-01-27
The Exact Pages split of CorpFin 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 |
|---|---|
| Kimi K3 | 69.3 |
| Inkling | 68.9 |
| Claude Opus 5 | 68.6 |
| Inkling-Small | 68.6 |
| Gemini 3 Flash Preview | 68.1 |
| GPT-5.2 | 68.1 |
| Grok 4 Fast | 67.2 |
| Grok 3 Mini | 67.0 |
| Claude Fable 5 | 66.9 |
| Grok 4.3 | 66.8 |
| Muse Spark 1.1 | 66.4 |
| Grok 4.5 | 66.3 |
| Kimi K2.5 | 66.2 |
| Grok 4 | 66.0 |
| GPT-5.5 | 65.6 |
| Grok 4.1 Fast | 65.5 |
| Opus 4.6 | 64.9 |
| MiniMax M3 | 64.9 |
| Qwen3 Max Thinking (2026-01-23) | 64.9 |
| Kimi K2.6 | 64.8 |
| Muse Spark | 64.8 |
| Opus 4.7 | 64.3 |
| Opus 4.8 | 64.3 |
| GLM-4.5 | 64.3 |
| Nemotron 3 Ultra 550B A55B | 64.3 |
| GPT-5.4 | 64.1 |
| O4 Mini | 64.0 |
| Sonnet 5 | 63.9 |
| GLM-5.2 | 63.6 |
| GPT-5.6 Luna | 63.5 |
| o3 | 63.5 |
| Opus 4.5 | 63.4 |
| Qwen3.6-Max-Preview | 63.3 |
| Sonnet 4.6 | 63.2 |
| GPT-5.6 Terra | 63.2 |
| GPT-5.1 | 63.1 |
| Grok 4.20 | 63.1 |
| Gemini 2.5 Pro | 62.9 |
| Gemini 3.5 Flash | 62.8 |
| GPT-4.1 | 62.8 |