ProLLM - LLM-as-a-Judge: leaderboard
ProLLM benchmark evaluating how well LLMs can judge and score the outputs of other AI models - measuring meta-evaluation capability.
Metric: Score (%). Source: www.prollm.ai. Saturation forecast: Around January 2027. 58 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4.1 | 85.3 | #248 |
| 2 | GPT-4o | 84.6 | #343 |
| 3 | GPT-OSS-120B | 84.6 | #341 |
| 4 | GPT-4 Turbo | 83.8 | #404 |
| 5 | O1 Mini | 83.8 | #323 |
| 6 | DeepSeek V3 | 82.4 | #321 |
| 7 | GPT-4.5 | 82.4 | #274 |
| 8 | O1 | 82.4 | #183 |
| 9 | O3 Mini (Medium) | 82.4 | #272 (O3 Mini) |
| 10 | O3 Mini (High) | 81.6 | #272 (O3 Mini) |
| 11 | Qwen 2.5 72B Instruct | 80.9 | #454 |
| 12 | Llama 4 Maverick Instruct | 80.9 | #456 |
| 13 | GPT-4.1 Mini | 80.1 | #355 |
| 14 | Llama 3.3 70B Instruct | 79.4 | #534 |
| 15 | GPT-OSS-20B | 79.4 | #518 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=prollm-llm-as-a-judge · How It Works · Data refreshed daily, snapshot 2026-10-20.