Leaderboard bars
19 ranked models · higher is better
View accessible chart data
| Model | Rank | Provider | Score |
|---|---|---|---|
| Kimi K3 | #1 | Moonshot AI | 52.7% |
| Grok 4.5 | #2 | xAI | 51.4% |
| GPT-5.6 Sol | #3 | OpenAI | 51.2% |
| Claude Fable 5 | #4 | Anthropic | 48.6% |
| Claude Opus 4.8 | #5 | Anthropic | 48.5% |
| GPT-5.6 Terra | #6 | OpenAI | 45.6% |
| Muse Spark 1.1 | #7 | Meta | 42.8% |
| Gemini 3.5 Flash | #8 | 42.6% | |
| GPT-5.6 Luna | #9 | OpenAI | 42.2% |
| GPT-5.5 | #10 | OpenAI | 42.1% |
| Claude Sonnet 5 | #11 | Anthropic | 39.2% |
| Gemini 3.1 Pro Preview | #12 | 37.5% | |
| GLM-5.2 | #13 | Z.AI | 27.8% |
| Qwen3.7-Max | #14 | Alibaba Cloud | 25.6% |
| Kimi K2.7 Code | #15 | Moonshot AI | 22.5% |
| Kimi K2.6 | #16 | Moonshot AI | 19.6% |
| MiMo-V2-Flash | #17 | Xiaomi | 18.8% |
| Qwen3.5 397B | #18 | Alibaba Cloud | 12.2% |
| Nemotron 3 Ultra | #19 | NVIDIA | 5.7% |

