Mercor GDPval Extended: leaderboard
Models complete professional-work tasks from Mercor’s held-out GDPval extension. The native mean rubric score stays separate from the original public benchmark and from all-pass task success.
Metric: Mean rubric score (%). Source: www.mercor.com. 44 models tracked.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | Gemini 4 Argon | High | 76.4 | #14 |
| 2 | Claude Opus 5.5 | Max | 75.8 | #7 |
| 3 | Claude Sonnet 5.5 | Max | 70.9 | #21 |
| 4 | Muse Wonder | Max | 70 | #31 |
| 5 | Claude Fable 5.1 | Max | 68.7 | #17 |
| 6 | Claude Haiku 5.5 | Max | 66 | #76 |
| 7 | Claude Fable 5 | Max | 65.3 | #28 |
| 8 | GPT-6 Astra | Max | 64.1 | #8 |
| 9 | Claude Opus 5 | Max | 62.9 | #23 |
| 10 | Grok 4.6 | xHigh | 61.8 | #47 |
| 11 | Gemini 3.8 Flash | High | 60.4 | #39 |
| 12 | Muse Spark Open | xHigh | 59.9 | #81 |
| 13 | Claude Opus 4.8 | Max | 58.9 | #62 |
| 14 | GPT-6 Sol | Max | 58.7 | #29 |
| 15 | DeepSeek V4.1 Flash | Max | 58.6 | #69 |
No result here: #1 Claude Fable 5.2, #2 GPT-6.2 Astra, #3 Gemini 4 Carbon.
Interactive version: theaggregate.ai/benchmark?slug=mercor-gdpval-extended · How It Works · Data refreshed daily, snapshot 2026-11-17.