Model rankings
1022 models · 5 boards · 2 sources
Most preferred
Best open model
Best from China
Most voted
Top three by Score
Gemini 4 Carbon (High)Google1,528
Gemini 4 Argon (High)Google1,520
Claude Opus 5.5 (High)Anthropic1,509
Top three open models
Kimi K3 (Max)Moonshot AI · #16 overall1,488
GLM-5.3 (Max)Z.ai · #25 overall1,481
MiMo-V2.6-ProXiaomi · #26 overall1,480
This week's changes
HeyGen Video 1Video · #3New
Kimi K3.1 ProIntelligence · #4New
Gemini 4 Carbon (High)Coding · #4New
GPT-6.1 Terra (Max)Coding · #13New
MiniMax-M3.1-FlashIntelligence · #18New
| Rank | Change | Model | Score | 95% interval | Votes | Price $/M |
|---|---|---|---|---|---|---|
| 1 | — | Kimi K3 (Max)OpenMoonshot AI | 1,488 | ±4 | 36,780 | $6.0 |
| 2 | — | Qwen3.8-MaxClosedAlibaba | 1,482 | ±5 | 31,915 | $3.0 |
| 3 | — | GLM-5.3 (Max)OpenZ.ai | 1,481 | ±5 | 26,606 | $2.2 |
| 4 | — | MiMo-V2.6-ProOpenXiaomi | 1,480 | ±9 | 4,037 | $0.54 |
| 5 | ▲ 1 | GLM-5.3-FlashOpenZ.ai | 1,477 | ±5 | 41,642 | $0.24 |
| 6 | ▼ 1 | DeepSeek-V4.1-Flash-20260910 (Max)OpenDeepSeek | 1,476 | ±6 | 13,657 | — |
| 7 | — | GLM-5.2 (Max)OpenZ.ai | 1,475 | ±4 | 53,702 | — |
| 8 | — | Qwen3.7-Max-PreviewClosedAlibaba | 1,474 | ±10 | 3,932 | — |
| 9 | — | ERNIE 5.1ClosedBaidu | 1,467 | ±5 | 39,469 | — |
| 10 | — | MiMo-V2.5-ProOpenXiaomi | 1,467 | ±4 | 77,728 | — |
| 11 | — | GLM-5.1OpenZ.ai | 1,465 | ±4 | 66,771 | — |
| 12 | ▲ 1 | DeepSeek-V4-Pro (High, 20260813)OpenDeepSeek · 1 more variant | 1,465 | ±7 | 9,787 | — |
| 13 | ▼ 1 | Qwen3.5-Max-PreviewClosedAlibaba | 1,465 | ±5 | 22,582 | — |
| 14 | — | Kimi K2.6OpenMoonshot AI | 1,461 | ±4 | 39,978 | — |
| 15 | — | Qwen3.6-Max-PreviewClosedAlibaba | 1,460 | ±8 | 5,419 | — |
| 16 | — | GLM-5OpenZ.ai | 1,457 | ±4 | 29,255 | — |
| 17 | — | Dola Seed 2.0 ProClosedByteDance | 1,456 | ±3 | 79,094 | — |
| 18 | — | Qwen3.7-PlusClosedAlibaba | 1,456 | ±4 | 42,011 | — |
| 19 | — | DeepSeek-V4-Pro Preview (High)OpenDeepSeek | 1,455 | ±4 | 54,870 | — |
| 20 | — | Hy3OpenTencent | 1,454 | ±6 | 13,840 | — |
| 21 | — | Kimi K2.5 (Thinking)OpenMoonshot AI · 1 more variant | 1,450 | ±3 | 74,682 | — |
| 22 | — | ERNIE-5.0-Preview (1203)ClosedBaidu · 1 more variant | 1,448 | ±6 | 9,802 | — |
| 23 | ▲ 2 | Step 5 PreviewClosedStepFun | 1,448 | ±6 | 11,355 | $1.4 |
| 24 | — | MiMo-V2-ProClosedXiaomi | 1,448 | ±5 | 25,421 | — |
| 25 | ▼ 2 | MiMo-V2.6-FlashOpenXiaomi | 1,448 | ±6 | 15,312 | $0.18 |
| 26 | — | ERNIE 5.0 (0110)ClosedBaidu | 1,446 | ±4 | 36,475 | — |
| 27 | — | Qwen3.6-PlusClosedAlibaba | 1,443 | ±4 | 48,126 | — |
| 28 | — | Qwen3.5-397B-A17BOpenAlibaba | 1,441 | ±3 | 97,675 | — |
| 29 | — | GLM-4.7OpenZ.ai | 1,441 | ±6 | 12,239 | — |
| 30 | — | MiniMax-M3OpenMiniMax | 1,439 | ±4 | 68,611 | $0.52 |
| 31 | — | DeepSeek-V4-Flash Preview (High)OpenDeepSeek | 1,439 | ±4 | 51,763 | — |
| 32 | ▲ 1 | LongCat-Flash-Chat Exp (2602)ClosedMeituan | 1,436 | ±5 | 29,244 | — |
| 33 | ▲ 1 | DeepSeek-V4-FlashOpenDeepSeek | 1,436 | ±4 | 51,944 | — |
| 34 | ▼ 2 | Qwen3.8-27BOpenAlibaba | 1,436 | ±5 | 27,779 | $1.1 |
| 35 | — | Qwen3-Max-PreviewClosedAlibaba | 1,434 | ±5 | 27,368 | — |
| 36 | ▲ 1 | GLM-5V-TurboClosedZ.ai | 1,434 | ±6 | 10,476 | — |
| 37 | ▼ 1 | MiMo-V2.5OpenXiaomi | 1,434 | ±4 | 47,327 | — |
| 38 | — | MiMo-V2-OmniClosedXiaomi | 1,431 | ±6 | 20,615 | — |
| 39 | — | Kimi K2 Thinking TurboOpenMoonshot AI | 1,430 | ±3 | 63,310 | — |
| 40 | — | DeepSeek-V3.2-Exp (Thinking)OpenDeepSeek · 1 more variant | 1,425 | ±7 | 8,991 | — |
| 41 | — | DeepSeek-V3.2OpenDeepSeek · 1 more variant | 1,425 | ±3 | 48,095 | — |
| 42 | — | GLM-4.6OpenZ.ai | 1,424 | ±4 | 35,810 | — |
| 43 | — | Qwen3-Max (2025-09-23)ClosedAlibaba | 1,423 | ±6 | 9,026 | — |
| 44 | — | Qwen3-235B-A22B-Instruct (2507)OpenAlibaba | 1,422 | ±3 | 98,081 | — |
| 45 | — | DeepSeek-R1 (0528)OpenDeepSeek · 1 more variant | 1,421 | ±6 | 18,165 | — |
| 46 | — | Kimi K2 Preview (0905)OpenMoonshot AI · 1 more variant | 1,418 | ±7 | 11,592 | — |
| 47 | — | DeepSeek-V3.1OpenDeepSeek · 1 more variant | 1,417 | ±6 | 14,606 | — |
| 48 | — | Qwen3.5-122B-A10BOpenAlibaba | 1,416 | ±4 | 29,712 | — |
| 49 | — | DeepSeek-V3.1-Terminus (Thinking)OpenDeepSeek · 1 more variant | 1,416 | ±10 | 3,379 | — |
| 50 | — | MiniMax-M2.7OpenMiniMax | 1,415 | ±3 | 89,197 | — |
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