Model rankings
769 models · 5 boards · 2 sources
Smartest
Best open model
Best from China
Top three by Intelligence
GPT-5.2 (xhigh)OpenAI50.6
Claude Opus 4.5 (Reasoning)Anthropic49.1
Gemini 3 Pro Preview (high)Google47.9
Top three open models
GLM-4.7 (Reasoning)Z AI · #9 overall41.7
DeepSeek V3.2 (Reasoning)DeepSeek · #12 overall41.2
Kimi K2 ThinkingKimi · #14 overall40.3
This week's changes
gpt-5.2-codexCoding · #17New
FLUX.2 [klein] 9BImage · #21New
FLUX.2 [klein] Base 9BImage · #43New
FLUX.2 [klein] 4BImage · #48New
GLM-ImageImage · #49New
| Rank | Change | Model | Intelligence | Price $/M | Speed tok/s | First token |
|---|---|---|---|---|---|---|
| 1 | — | GLM-4.7 (Reasoning)OpenZ AI · 1 more variant | 41.7 | — | — | — |
| 2 | — | DeepSeek V3.2 (Reasoning)OpenDeepSeek · 1 more variant | 41.2 | — | — | — |
| 3 | — | Kimi K2 ThinkingOpenKimi | 40.3 | — | — | — |
| 4 | — | MiniMax-M2.1OpenMiniMax | 39.4 | — | — | — |
| 5 | — | MiMo-V2-Flash (Reasoning)OpenXiaomi · 1 more variant | 39.1 | — | — | — |
| 6 | — | MiniMax-M2OpenMiniMax | 35.7 | — | — | — |
| 7 | — | DeepSeek V3.1 Terminus (Reasoning)OpenDeepSeek · 1 more variant | 33.5 | — | — | — |
| 8 | — | DeepSeek V3.2 Exp (Reasoning)OpenDeepSeek · 1 more variant | 32.5 | — | — | — |
| 9 | — | GLM-4.6 (Reasoning)OpenZ AI · 1 more variant | 32.2 | — | — | — |
| 10 | — | Qwen3 235B A22B 2507 (Reasoning)OpenAlibaba · 2 more variants | 29.4 | — | — | — |
| 11 | — | DeepSeek V3.2 SpecialeOpenDeepSeek | 29.0 | — | — | — |
| 12 | — | Kimi K2 0905OpenKimi | 28.0 | — | — | — |
| 13 | — | DeepSeek V3.1 (Reasoning)OpenDeepSeek · 1 more variant | 27.9 | — | — | — |
| 14 | — | Qwen3 VL 235B A22B (Reasoning)OpenAlibaba | 27.5 | — | — | — |
| 15 | — | DeepSeek R1 0528 (May '25)OpenDeepSeek · 1 more variant | 27.1 | — | — | — |
| 16 | — | GLM-4.5 (Reasoning)OpenZ AI | 26.8 | — | — | — |
| 17 | — | Qwen3 Next 80B A3B (Reasoning)OpenAlibaba | 26.5 | — | — | — |
| 18 | — | Seed-OSS-36B-InstructOpenByteDance Seed | 25.0 | — | — | — |
| 19 | — | Qwen3 Coder 480B A35B InstructOpenAlibaba | 24.8 | — | — | — |
| 20 | — | Qwen3 235B A22B 2507 InstructOpenAlibaba | 24.6 | — | — | — |
| 21 | — | MiniMax M1 80kOpenMiniMax | 24.3 | — | — | — |
| 22 | — | GLM-4.6V (Reasoning)OpenZ AI · 1 more variant | 23.8 | — | — | — |
| 23 | — | GLM-4.5-AirOpenZ AI | 23.6 | — | — | — |
| 24 | — | Qwen3 30B A3B 2507 (Reasoning)OpenAlibaba · 2 more variants | 22.8 | — | — | — |
| 25 | — | DeepSeek V3 0324OpenDeepSeek · 1 more variant | 22.6 | — | — | — |
| 26 | — | Qwen3 VL 235B A22B InstructOpenAlibaba | 20.8 | — | — | — |
| 27 | — | Qwen3 Next 80B A3B InstructOpenAlibaba | 20.3 | — | — | — |
| 28 | — | Qwen3 Coder 30B A3B InstructOpenAlibaba | 20.2 | — | — | — |
| 29 | — | Ling-1TOpenInclusionAI | 19.7 | — | — | — |
| 29 | — | Qwen3 VL 30B A3B (Reasoning)OpenAlibaba | 19.7 | — | — | — |
| 31 | — | Qwen3 4B 2507 (Reasoning)OpenAlibaba | 18.8 | — | — | — |
| 32 | — | Qwen3 VL 32B InstructOpenAlibaba | 17.1 | — | — | — |
| 33 | — | Qwen3 32B (Reasoning)OpenAlibaba | 17.1 | — | — | — |
| 34 | — | Qwen3 VL 8B (Reasoning)OpenAlibaba | 17.0 | — | — | — |
| 35 | — | Qwen3 14B (Reasoning)OpenAlibaba · 1 more variant | 16.8 | — | — | — |
| 36 | — | Qwen3 VL 30B A3B InstructOpenAlibaba | 16.3 | — | — | — |
| 37 | — | Qwen3 Omni 30B A3B (Reasoning)OpenAlibaba | 15.8 | — | — | — |
| 38 | — | Qwen3 30B A3B 2507 InstructOpenAlibaba | 15.4 | — | — | — |
| 39 | — | Qwen3 VL 8B InstructOpenAlibaba | 14.6 | — | — | — |
| 40 | — | GLM-4.5V (Non-reasoning)OpenZ AI | 13.2 | — | — | — |
| 41 | — | Qwen3 4B 2507 InstructOpenAlibaba | 13.1 | — | — | — |
| 42 | — | Qwen3 8B (Non-reasoning)OpenAlibaba | 10.8 | — | — | — |
| 43 | — | Qwen3 Omni 30B A3B InstructOpenAlibaba | 10.7 | — | — | — |
| 44 | — | Qwen3 1.7B (Non-reasoning)OpenAlibaba | 6.8 | — | — | — |
| 45 | — | Qwen3 0.6B (Non-reasoning)OpenAlibaba | 6.1 | — | — | — |
Intelligence over time
Gained most · Anthropic +57.2Highest now · Anthropic 64.6Gap between highest and lowest 18.1 ptsTime range Nov 2025 – Nov 2026
64.6Anthropic60.0Google56.9OpenAI50.1Meta46.5Kimi