AI Leaderboard
Which AI models lead right now: Arena's daily ranking for agents, chat, code, image and video.
Rankings across agent behavior signalsTop 25 of 53 modelsFull ranking on arena.ai
| Rank | Model | Net improvement | Confirmed success | License |
|---|---|---|---|---|
| 1could rank 1–8 | Claude Opus 5.5 (Max)Anthropic · Proprietary | +15.53%±4.63% | +17.29%±6.02% | Proprietary |
| 2could rank 1–7 | Claude Opus 5.5 (High)Anthropic · Proprietary | +14.33%±2.18% | +13.93%±3.84% | Proprietary |
| 3could rank 1–8 | GPT 6 Astra (Max)OpenAI · Proprietary | +13.09%±2.37% | +12.13%±3.88% | Proprietary |
| 4could rank 1–8 | Claude Fable 5.1 (Max)Anthropic · Proprietary | +12.66%±2.05% | +16.19%±2.96% | Proprietary |
| 5could rank 1–11 | Claude Sonnet 5.5 (Max)Anthropic · Proprietary | +11.95%±3.31% | +15.83%±5.06% | Proprietary |
| 6could rank 1–11 | GPT 6.1 Sol (Max)OpenAI · Proprietary | +11.72%±3.05% | +14.74%±4.18% | Proprietary |
| 7could rank 1–12 | GPT 6 Sol (Max)OpenAI · Proprietary | +9.92%±2.38% | +6.97%±4.34% | Proprietary |
| 8could rank 2–13 | Gemini 4 Argon (High)Google · Proprietary | +9.26%±2.36% | +15.64%±3.26% | Proprietary |
| 9could rank 5–12 | Claude Fable 5 (High)Anthropic · Proprietary | +8.94%±1.27% | +5.79%±2.60% | Proprietary |
| 10could rank 5–13 | Claude Opus 5 (Max)Anthropic · Proprietary | +8.12%±1.47% | +10.42%±3.10% | Proprietary |
| 11could rank 5–14 | Claude Opus 5 (High)Anthropic · Proprietary | +7.95%±1.39% | +8.37%±2.99% | Proprietary |
| 12could rank 7–14 | Claude Opus 4.8 (High)Anthropic · Proprietary | +7.13%±1.37% | +4.68%±2.75% | Proprietary |
| 13could rank 9–16 | GPT 5.6 Sol (xHigh)OpenAI · Proprietary | +6.23%±1.28% | +5.34%±2.70% | Proprietary |
| 14could rank 11–23 | Claude Sonnet 5 (High)Anthropic · Proprietary | +4.91%±1.69% | +2.22%±3.48% | Proprietary |
| 15could rank 13–23 | GPT 5.5 (xHigh)OpenAI · Proprietary | +4.38%±1.09% | -0.88%±2.34% | Proprietary |
| 16could rank 14–23 | Muse Spark 1.3 (Max)Meta · Proprietary | +4.04%±0.66% | +5.32%±1.39% | Proprietary |
| 17could rank 14–23 | Deepseek V4.1 Flash (Max)DeepSeek · MIT | +3.80%±0.39% | +7.44%±0.86% | MIT |
| 18could rank 14–23 | Kimi K3 (Max)Moonshot · Kimi K3 license | +3.78%±0.59% | +6.46%±1.32% | Kimi K3 license |
| 19could rank 14–25 | Hy4 previewTencent · Apache 2.0 | +3.61%±0.70% | +5.96%±1.48% | Apache 2.0 |
| 20could rank 13–33 | MiMo V2.6 ProXiaomi · MIT | +3.52%±1.92% | +8.03%±3.88% | MIT |
| 21could rank 14–27 | GLM 5.2 (Max)Z.ai · MIT | +3.19%±0.73% | +4.11%±1.58% | MIT |
| 22could rank 14–34 | Grok 4.7 (xHigh)SpaceXAI · Proprietary | +2.83%±1.71% | +3.93%±3.66% | Proprietary |
| 23could rank 14–33 | Gemini 3.8 Flash (High)Google · Proprietary | +2.75%±0.93% | +5.30%±2.05% | Proprietary |
| 24could rank 19–33 | Qwen3.8 MaxAlibaba · Proprietary | +2.35%±0.59% | +5.09%±1.32% | Proprietary |
| 25could rank 20–34 | GLM 5.3 (Max)Z.ai · MIT | +2.15%±0.72% | +4.78%±1.62% | MIT |