Model rankings
517 models · 3 boards · 2 sources
Smartest
Best open model
Best value
Best from China
Top three open models
Qwen3 235B A22B 2507 (Reasoning)Alibaba · #8 overall63.6
gpt-oss-120BOpenAI · #9 overall61.3
DeepSeek V3.1 (Reasoning)DeepSeek · #11 overall59.7
This week's changes
DeepSeek V3.1 (Reasoning)Intelligence · #11New
NVIDIA Nemotron Nano 9B V2 (Reasoning)Intelligence · #44New
GLM-4.5V (Reasoning)Intelligence · #45New
LFM2 1.2BIntelligence · #164New
Llama 3.2 Instruct 3BIntelligence · #152Up 26
| Rank | Change | Model | Intelligence | Price $/M | Speed tok/s | First token |
|---|---|---|---|---|---|---|
| 1 | — | GPT-5 (high)ClosedOpenAI | 69.0 | $3.4 | — | — |
| 2 | — | GPT-5 (medium)ClosedOpenAI | 67.5 | $3.4 | — | — |
| 3 | — | Grok 4ClosedxAI | 67.5 | $6.0 | — | — |
| 4 | — | o3-proClosedOpenAI | 67.5 | $35 | — | — |
| 5 | — | o3ClosedOpenAI | 67.1 | $3.5 | — | — |
| 6 | New | GPT-5 mini (high)ClosedOpenAI | 65.4 | $0.000 | — | — |
| 7 | ▼ 1 | o4-mini (high)ClosedOpenAI | 65.1 | $1.9 | — | — |
| 8 | ▼ 1 | Gemini 2.5 ProClosedGoogle | 64.6 | $3.4 | — | — |
| 9 | New | GPT-5 mini (medium)ClosedOpenAI | 63.7 | $0.69 | — | — |
| 10 | ▼ 1 | GPT-5 (low)ClosedOpenAI | 63.4 | $3.4 | — | — |
| 11 | ▼ 1 | gpt-oss-120BOpenOpenAI | 61.3 | $0.26 | — | — |
| 12 | ▼ 1 | Claude 4.1 Opus (Extended Thinking)ClosedAnthropic | 61.0 | $30 | — | — |
| 13 | ▼ 1 | Gemini 2.5 Pro Preview (Mar' 25)ClosedGoogle | 59.0 | $3.4 | — | — |
| 14 | ▼ 1 | Claude 4 Sonnet (Extended Thinking)ClosedAnthropic | 58.8 | $6.0 | — | — |
| 15 | ▼ 1 | Gemini 2.5 Flash (Reasoning)ClosedGoogle | 58.4 | $0.85 | — | — |
| 16 | ▼ 1 | Grok 3 mini Reasoning (high)ClosedxAI | 58.0 | $0.35 | — | — |
| 17 | ▼ 1 | Gemini 2.5 Pro Preview (May' 25)ClosedGoogle | 58.0 | $3.4 | — | — |
| 18 | ▼ 1 | o3-mini (high)ClosedOpenAI | 55.5 | $1.9 | — | — |
| 19 | ▼ 1 | Claude 4 Opus (Extended Thinking)ClosedAnthropic | 55.3 | $30 | — | — |
| 20 | New | GPT-5 nano (high)ClosedOpenAI | 54.7 | $0.000 | — | — |
| 21 | New | GPT-5 nano (medium)ClosedOpenAI | 53.8 | $0.14 | — | — |
| 22 | ▼ 2 | o3-miniClosedOpenAI | 52.7 | $1.9 | — | — |
| 23 | ▼ 2 | Llama Nemotron Super 49B v1.5 (Reasoning)OpenNVIDIA | 52.4 | $0.000 | — | — |
| 24 | ▼ 2 | o1ClosedOpenAI | 51.7 | $26 | — | — |
| 25 | ▼ 2 | Sonar Reasoning ProClosedPerplexity | 50.7 | $0.000 | — | — |
| 26 | ▼ 2 | Gemini 2.5 Flash Preview (Reasoning)ClosedGoogle | 50.2 | $0.99 | — | — |
| 27 | ▼ 2 | o1-previewClosedOpenAI | 49.3 | $29 | — | — |
| 28 | ▼ 2 | gpt-oss-20BOpenOpenAI | 49.0 | $0.088 | — | — |
| 29 | ▼ 2 | Claude 4.1 OpusClosedAnthropic | 49.0 | $30 | — | — |
| 30 | ▼ 2 | Gemini 2.5 FlashClosedGoogle | 47.3 | $0.85 | — | — |
| 31 | ▼ 2 | Gemini 2.5 Flash-Lite (Reasoning)ClosedGoogle | 47.3 | $0.18 | — | — |
| 32 | ▼ 2 | Claude 3.7 Sonnet (Extended Thinking)ClosedAnthropic | 47.0 | $6.0 | — | — |
| 33 | ▼ 2 | GPT-4.1ClosedOpenAI | 46.8 | $3.5 | — | — |
| 34 | ▼ 2 | Claude 4 OpusClosedAnthropic | 46.6 | $30 | — | — |
| 35 | ▼ 2 | Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)OpenNVIDIA | 46.4 | $0.90 | — | — |
| 36 | New | GPT-5 mini (minimal)ClosedOpenAI | 46.3 | $0.000 | — | — |
| 37 | ▼ 3 | Claude 4 SonnetClosedAnthropic | 45.7 | $6.0 | — | — |
| 38 | ▼ 3 | Grok 3 Reasoning BetaClosedxAI | 45.6 | $0.000 | — | — |
| 38 | ▼ 3 | o1-proClosedOpenAI | 45.6 | $263 | — | — |
| 40 | ▼ 3 | GPT-4.1 miniClosedOpenAI | 45.6 | $0.70 | — | — |
| 41 | ▼ 3 | GPT-5 (minimal)ClosedOpenAI | 43.8 | $3.4 | — | — |
| 42 | ▼ 3 | o1-miniClosedOpenAI | 43.3 | $1.9 | — | — |
| 43 | New | NVIDIA Nemotron Nano 9B V2 (Reasoning)OpenNVIDIA | 43.2 | $0.000 | — | — |
| 44 | ▼ 4 | GPT-4.5 (Preview)ClosedOpenAI | 42.4 | $0.000 | — | — |
| 45 | ▼ 4 | Llama 4 MaverickOpenMeta | 41.7 | $0.39 | — | — |
| 46 | ▼ 4 | Gemini 2.0 Flash Thinking Experimental (Jan '25)ClosedGoogle | 41.7 | $0.000 | — | — |
| 47 | ▼ 3 | Grok 3ClosedxAI | 39.9 | $6.0 | — | — |
| 48 | ▼ 3 | GPT-4o (March 2025, chatgpt-4o-latest)ClosedOpenAI | 39.5 | $7.5 | — | — |
| 49 | ▼ 6 | Llama 3.3 Nemotron Super 49B v1 (Reasoning)OpenNVIDIA | 39.4 | $0.000 | — | — |
| 50 | ▼ 4 | Gemini 2.0 Pro Experimental (Feb '25)ClosedGoogle | 38.4 | $0.000 | — | — |
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