MMLU-CF 0-shot Test: leaderboard
MMLU-CF 0-shot Test — Official MMLU-CF contamination-free benchmark; 0-shot test cohort. Original MMLU column and test-minus-validation deltas are not independent capability panels. Native API/checkpoint labels retained. Metric: Accuracy (%). Source: https://github.com/microsoft/MMLU-CF
Metric: Accuracy (%). Source: github.com. Saturation forecast: Estimated already saturated. 44 models tracked. Updated 16 Nov 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4o | 71.9 | #744 |
| 2 | Qwen 2.5 72B Instruct | 70.6 | #1095 |
| 3 | Llama 3.1 70B Instruct | 70.4 | #1542 |
| 4 | GPT-4 Turbo | 68.9 | #999 |
| 5 | Qwen 2.5 32B Instruct | 68.9 | #1232 |
| 6 | phi-4-14B | 68.5 | #1666 |
| 7 | Llama 3 70B Instruct | 68.1 | #1689 |
| 8 | Llama 3.3 70B Instruct | 67.6 | #1379 |
| 9 | Qwen 2.5 14B Instruct | 67 | #1750 |
| 10 | GPT-4o Mini | 66 | #1658 |
| 11 | Mixtral 8x22B Instruct | 65.3 | #2097 |
| 12 | Gemma 2 27B | 64.2 | #2141 |
| 13 | Phi-3.5-MoE-instruct | 63.1 | #2199 |
| 14 | Phi-3-medium-128k-instruct | 62.5 | #3321 |
| 15 | Qwen 2 72B Instruct | 62.4 | #1514 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=mmlu-cf-0-shot-test · How It Works · Data refreshed daily, snapshot 2026-11-16.