| # | Model | Org | Status | MMLU | Arena ELO | Δ 7d | Trend |
|---|---|---|---|---|---|---|---|
| 1 | Claude Opus 4.6 | Anthropic | LIVE | 92.8 | 1,412 | +4.1 | |
| 2 | GPT-5 Turbo | OpenAI | LIVE | 92.1 | 1,402 | +3.2 | |
| 3 | Gemini Ultra 2.5 | Google DeepMind | LIVE | 90.2 | 1,375 | -0.4 | |
| 4 | DeepSeek-V4 | DeepSeek | LIVE | 89.1 | 1,355 | +4.3 | |
| 5 | Llama 5 70B | Meta AI | BETA | 88.5 | 1,340 | +5.1 | |
| 6 | Mistral Large 3 | Mistral AI | NEW | 86.9 | 1,310 | +0.9 | |
| 7 | Grok-3 | xAI | LIVE | 87.2 | 1,325 | +2.8 |
| 8 | Command R+ Ultra | Cohere | LIVE | 84.2 | 1,275 | -1.1 | |
| 9 | Yi-Lightning | 01.AI | LIVE | 83.8 | 1,260 | +1.4 | |
| 10 | Qwen3-72B | Alibaba | BETA | 83.5 | 1,248 | +2.0 |