Leaderboard bars
20 ranked models · higher is better
View accessible chart data
| Model | Rank | Provider | Score |
|---|---|---|---|
| GPT-5.6 Sol | #1 | OpenAI | 56.2% |
| GPT-5.6 Terra | #2 | OpenAI | 51.0% |
| Claude Opus 4.7 | #3 | Anthropic | 46.7% |
| GPT-5.5 | #4 | OpenAI | 45.8% |
| GLM-5.2 | #5 | Z.AI | 42.7% |
| Qwen3.7-Max | #6 | Alibaba Cloud | 42.5% |
| Gemini 3.5 Flash | #7 | 40.3% | |
| GPT-5.6 Luna | #8 | OpenAI | 40.3% |
| GLM-5.1 | #9 | Z.AI | 40.3% |
| Claude Sonnet 4.6 | #10 | Anthropic | 39.8% |
| DeepSeek V4 Pro | #11 | DeepSeek | 38.3% |
| MiMo-V2.5-Pro | #12 | Xiaomi | 38.2% |
| Gemma 4 31B | #13 | 37.3% | |
| Qwen3.5 27B | #14 | Alibaba Cloud | 35.5% |
| GPT-5.4 mini | #15 | OpenAI | 35.2% |
| Qwen3.5 397B | #16 | Alibaba Cloud | 34.1% |
| Grok 4.3 | #17 | xAI | 32.7% |
| DeepSeek V4 Flash | #18 | DeepSeek | 31.5% |
| Kimi K2.6 | #19 | Moonshot AI | 31.2% |
| Gemini 3.1 Pro Preview | #20 | 30.3% |

