Artificial Analysis Harvey LAB-AA leaderboard
19 ranked models · higher is better
View accessible chart data
| Model | Rank | Provider | Score |
|---|---|---|---|
| Kimi K3 | #1 | Moonshot AI | 94.6% |
| Claude Fable 5 | #2 | Anthropic | 93.6% |
| Muse Spark 1.1 | #3 | Meta | 93.1% |
| Grok 4.5 | #4 | xAI | 92.4% |
| Claude Opus 4.8 | #5 | Anthropic | 91.1% |
| GLM-5.2 | #6 | Z.AI | 91% |
| Claude Sonnet 5 | #7 | Anthropic | 90.1% |
| MiniMax M3 | #8 | MiniMax | 88.4% |
| GPT-5.6 Luna | #9 | OpenAI | 87.9% |
| GPT-5.6 Sol | #10 | OpenAI | 87.2% |
| GPT-5.5 | #11 | OpenAI | 86.3% |
| GPT-5.6 Terra | #12 | OpenAI | 85.2% |
| DeepSeek V4 Pro (Max) | #13 | DeepSeek | 84.4% |
| Qwen3.7-Max | #14 | Alibaba Cloud | 83.4% |
| Nemotron 3 Ultra | #15 | NVIDIA | 81.7% |
| DeepSeek V4 Flash (Max) | #16 | DeepSeek | 81.3% |
| MiMo-V2.5-Pro | #17 | Xiaomi | 73.3% |
| Mistral Medium 3.5 128B | #18 | Mistral AI | 69.1% |
| Gemini 3.1 Pro | #19 | 58.9% |