Artificial Analysis Terminal-Bench v2.1 leaderboard
17 ranked models · higher is better
View accessible chart data
| Model | Rank | Provider | Score |
|---|---|---|---|
| GPT-5.6 Sol | #1 | OpenAI | 88% |
| GPT-5.6 Terra | #2 | OpenAI | 88% |
| Kimi K3 | #3 | Moonshot AI | 85% |
| Claude Fable 5 | #4 | Anthropic | 84.6% |
| Claude Opus 4.8 | #5 | Anthropic | 84.6% |
| GPT-5.5 | #6 | OpenAI | 84.3% |
| Grok 4.5 | #7 | xAI | 81.6% |
| GPT-5.6 Luna | #8 | OpenAI | 80.9% |
| Claude Sonnet 5 | #9 | Anthropic | 80.5% |
| GLM-5.2 | #10 | Z.AI | 77.9% |
| Muse Spark 1.1 | #11 | Meta | 77.9% |
| Gemini 3.6 Flash | #12 | 77.5% | |
| Qwen3.7-Max | #13 | Alibaba Cloud | 74.5% |
| Gemini 3.1 Pro | #14 | 73.8% | |
| MiMo-V2.5-Pro | #15 | Xiaomi | 65.2% |
| MiniMax M3 | #16 | MiniMax | 65.2% |
| DeepSeek V4 Pro (Max) | #17 | DeepSeek | 64% |