Model rankings
866 models · 5 boards · 2 sources
Smartest
Best open model
Best value
Best from China
Top three by Intelligence
Gemini 4 Carbon (High)Google59.4
Claude Opus 5.5 (Max, Default Fallback)Anthropic58.4
GPT-6 Astra (Max)OpenAI56.9
Top three open models
DeepSeek V4.1 Flash 1017 (Max)DeepSeek · #15 overall32.5
GLM-5.3 (Max)Z AI · #16 overall31.7
MiMo-V2.6-ProXiaomi · #18 overall29.9
This week's changes
Gemini 4 Carbon (High)Intelligence · #1New
GPT-6.1 Terra (Max)Intelligence · #6New
Fish Audio Drama 3Speech · #6New
P-Video-2-ProVideo · #11New
AssemblyAI-TTS-1Speech · #14New
| Rank | Change | Model | Intelligence | Price $/M | Speed tok/s | First token |
|---|---|---|---|---|---|---|
| 1 | ▲ 1 | DeepSeek V4.1 Flash 1017 (Max)OpenDeepSeek | 32.5 | $0.52 | 174 | 1.2 s |
| 2 | ▲ 1 | GLM-5.3 (Max)OpenZ AI | 31.7 | $2.2 | 90 | 1.4 s |
| 3 | ▼ 2 | MiMo-V2.6-ProOpenXiaomi | 29.9 | $0.54 | 44 | 3.5 s |
| 4 | New | Ling 3.1 Flash 1019OpenInclusionAI | 29.5 | $0.45 | — | — |
| 5 | New | MiniMax-M3.1-FlashOpenMiniMax | 28.9 | $0.26 | 120 | 0.9 s |
| 6 | New | DeepSeek V4.1 Flash 1017 (High)OpenDeepSeek | 28.5 | $0.52 | 218 | 1.0 s |
| 7 | ▼ 3 | Step 5 PreviewOpenStepFun | 28.5 | $1.4 | 142 | 0.9 s |
| 8 | ▼ 3 | GLM-5.3-FlashOpenZ AI | 27.8 | $0.24 | 74 | 1.5 s |
| 9 | ▲ 1 | MiMo-V2.6-FlashOpenXiaomi | 26.7 | $0.18 | 36 | 2.7 s |
| 10 | New | Mistral Large 4 (High)OpenMistral | 26.0 | $1.0 | — | — |
| 11 | ▼ 2 | Muse Spark Open (Xhigh)OpenMeta | 24.7 | $2.0 | 188 | 0.7 s |
| 12 | New | Mistral Large 4 (Medium)OpenMistral | 23.1 | $1.0 | — | — |
| 13 | ▼ 2 | DeepSeek V4 Pro 0813 (Max)OpenDeepSeek | 23.1 | $2.0 | 83 | 1.8 s |
| 14 | ▼ 2 | GLM-5.3 (Low)OpenZ AI | 23.1 | $2.2 | 84 | 1.7 s |
| 15 | New | Nanbeige4.3-33BOpenNanbeige | 20.4 | — | — | — |
| 16 | ▼ 3 | Qwen3.8 27B (Xhigh)OpenAlibaba | 20.1 | $1.1 | 41 | 4.0 s |
| 17 | — | MiniMax-M3OpenMiniMax | 18.7 | $0.52 | 112 | 0.8 s |
| 18 | New | DeepSeek V4.1 Flash 1017 (Non-reasoning)OpenDeepSeek | 18.5 | $0.52 | 185 | 1.4 s |
| 19 | ▼ 4 | K2 Horizon 375B A23BOpenInstitute of Foundation Models | 17.8 | $0.000 | 665 | 2.4 s |
| 20 | ▼ 4 | Beam 1.0OpenReflection | 17.7 | $0.60 | 357 | 0.7 s |
| 21 | ▼ 2 | Qwen3.8 27B (Medium)OpenAlibaba | 17.5 | $1.1 | 48 | 4.0 s |
| 22 | ▼ 1 | Qwen3.8 27B (Low)OpenAlibaba | 15.7 | $1.1 | 47 | 4.1 s |
| 23 | ▼ 3 | Nemotron 3.5 Super VLOpenNVIDIA | 13.6 | $0.28 | 464 | 0.7 s |
| 24 | ▼ 1 | Inkling SmallOpenThinking Machines | 12.9 | $0.52 | 226 | 1.6 s |
| 25 | ▲ 1 | Inkling (Xhigh)OpenThinking Machines | 12.5 | $1.8 | 211 | 1.7 s |
| 26 | ▲ 3 | Nemotron 3 Ultra 550B A55B (Reasoning)OpenNVIDIA | 11.8 | $1.1 | 163 | 1.2 s |
| 27 | ▲ 10 | Qwen3.8 27B (Non-reasoning)OpenAlibaba | 10.0 | $1.1 | 45 | 4.0 s |
| 28 | ▲ 15 | Muse Glimmer (High)OpenMeta | 6.3 | $0.52 | 126 | 1.1 s |
| 29 | ▲ 27 | Command A+OpenCohere | 6.2 | $0.000 | 176 | 0.5 s |
| 30 | ▲ 29 | Nemotron 3.5 LightningOpenNVIDIA | 5.5 | $0.095 | 295 | 0.6 s |
| 31 | ▲ 40 | Granite 4.2 8BOpenIBM | 5.1 | $0.11 | 61 | 1.0 s |
| 32 | ▲ 56 | Granite 4.2 3BOpenIBM | 4.7 | — | — | — |
| 33 | ▲ 35 | Mistral Small 4 (Reasoning)OpenMistral | 4.3 | $0.26 | 170 | 0.9 s |
| 34 | ▲ 137 | NVIDIA Nemotron 3 Nano 4BOpenNVIDIA | 1.8 | — | — | — |
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