ProfBench: leaderboard
Professional report-writing tasks graded against rubrics written by PhD physicists and chemists and MBA-level finance and consulting experts; rubric satisfaction across the four domains.
Metric: Overall Rubric Score (%). Source: huggingface.co. Saturation forecast: Around 2029. 102 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 5 (Thinking) | 66.2 | #12 (Claude Opus 5) |
| 2 | Claude Opus 4.7 (Thinking) | 61.3 | #49 (Claude Opus 4.7) |
| 3 | GPT-5.6 Sol (xHigh) | 60.4 | #19 (GPT-5.6 Sol) |
| 4 | Claude Sonnet 5 (Thinking) | 60.4 | #57 (Claude Sonnet 5) |
| 5 | Claude Opus 4.6 (Thinking) | 58.8 | #63 (Claude Opus 4.6) |
| 6 | GPT-5.5 (xHigh) | 58 | #31 (GPT-5.5) |
| 7 | Kimi K3 (Thinking) | 57.9 | #30 (Kimi K3) |
| 8 | GPT-5.6 Terra (xHigh) | 57.2 | #41 (GPT-5.6 Terra) |
| 9 | GPT-5.3 Codex (xHigh) | 57.2 | #69 (GPT-5.3 Codex) |
| 10 | GPT-5.4 (xHigh) | 56.8 | #79 (GPT-5.4) |
| 11 | Claude Opus 4.8 (Thinking) | 56.8 | #37 (Claude Opus 4.8) |
| 12 | Claude Sonnet 4.6 (Thinking) | 55.6 | #90 (Claude Sonnet 4.6) |
| 13 | GPT-5.6 Luna (xHigh) | 55.3 | #68 (GPT-5.6 Luna) |
| 14 | Gemini 3.1 Pro (Preview) | 55.2 | #58 |
| 15 | GPT-5.2 (xHigh) | 55.1 | #110 (GPT-5.2) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=profbench · How It Works · Data refreshed daily, snapshot 2026-10-20.