Model rankings
1190 models · 5 boards · 2 sources
Smartest
Best open model
Best value
Best from China
Top three by Intelligence
GPT-5.5 (xhigh)OpenAI60.2
Claude Opus 4.7 (Adaptive Reasoning, Max Effort)Anthropic57.3
Gemini 3.1 Pro PreviewGoogle57.2
Top three open models
Kimi K2.6Kimi · #5 overall53.9
MiMo-V2.5-ProXiaomi · #6 overall53.8
DeepSeek V4 Pro (Reasoning, Max Effort)DeepSeek · #13 overall51.5
This week's changes
gpt-5.5-high (codex-harness)Coding · #7New
Grok 4.3Intelligence · #8New
mimo-v2.5-proCoding · #9New
DeepSeek V4 Pro (Reasoning, Max Effort)Intelligence · #13New
gpt-5.5 (codex-harness)Coding · #15New
| Rank | Change | Model | Intelligence | Price $/M | Speed tok/s | First token |
|---|---|---|---|---|---|---|
| 1 | — | GPT-5.5 (xhigh)ClosedOpenAI · 4 more variants | 60.2 | $11 | 66 | 118 s |
| 2 | — | Claude Opus 4.7 (Adaptive Reasoning, Max Effort)ClosedAnthropic · 1 more variant | 57.3 | $10 | 49 | 26 s |
| 3 | — | Gemini 3.1 Pro PreviewClosedGoogle | 57.2 | $4.5 | 122 | 25 s |
| 4 | — | GPT-5.4 (xhigh)ClosedOpenAI · 2 more variants | 56.8 | $5.6 | 88 | 215 s |
| 5 | — | GPT-5.3 Codex (xhigh)ClosedOpenAI | 53.6 | $4.8 | 72 | 94 s |
| 6 | New | Grok 4.3ClosedxAI | 53.2 | $1.6 | 203 | 13 s |
| 7 | ▼ 1 | Claude Opus 4.6 (Adaptive Reasoning, Max Effort)ClosedAnthropic · 1 more variant | 53.0 | $10 | 46 | 23 s |
| 8 | ▼ 1 | Muse SparkClosedMeta | 52.2 | — | — | — |
| 9 | ▼ 1 | Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)ClosedAnthropic · 2 more variants | 51.7 | $6.0 | 60 | 117 s |
| 10 | ▼ 1 | GPT-5.2 (xhigh)ClosedOpenAI · 2 more variants | 51.3 | $4.8 | 68 | 159 s |
| 11 | ▼ 1 | Claude Opus 4.5 (Reasoning)ClosedAnthropic · 1 more variant | 49.7 | $10 | 50 | 16 s |
| 12 | ▼ 1 | Grok 4.20 0309 v2 (Reasoning)ClosedxAI · 1 more variant | 49.3 | $3.0 | 93 | 26 s |
| 13 | ▼ 1 | GPT-5.2 Codex (xhigh)ClosedOpenAI | 49.0 | $4.8 | 112 | 44 s |
| 14 | ▼ 1 | GPT-5.4 mini (xhigh)ClosedOpenAI · 2 more variants | 48.9 | $1.7 | 164 | 8.8 s |
| 15 | ▼ 1 | Grok 4.20 0309 (Reasoning)ClosedxAI · 1 more variant | 48.5 | $3.0 | 93 | 19 s |
| 16 | ▼ 1 | Gemini 3 Pro Preview (high)ClosedGoogle · 1 more variant | 48.4 | $4.5 | 128 | 28 s |
| 17 | ▼ 1 | GPT-5.1 (high)ClosedOpenAI · 1 more variant | 47.7 | $3.4 | 120 | 39 s |
| 18 | ▼ 1 | Gemini 3 Flash Preview (Reasoning)ClosedGoogle · 1 more variant | 46.4 | $1.1 | 165 | 7.7 s |
| 19 | ▼ 1 | GPT-5 (high)ClosedOpenAI · 4 more variants | 44.6 | $3.4 | 79 | 88 s |
| 19 | ▼ 1 | GPT-5 Codex (high)ClosedOpenAI | 44.6 | $3.4 | 176 | 7.4 s |
| 21 | ▼ 1 | GPT-5.4 nano (xhigh)ClosedOpenAI · 2 more variants | 44.0 | $0.46 | 160 | 5.0 s |
| 22 | ▼ 1 | GPT-5.1 Codex (high)ClosedOpenAI | 43.1 | $3.4 | 176 | 8.1 s |
| 23 | ▼ 1 | Claude 4.5 Sonnet (Reasoning)ClosedAnthropic · 1 more variant | 43.0 | $6.0 | 45 | 10 s |
| 24 | ▼ 1 | Claude 4.1 Opus (Reasoning)ClosedAnthropic · 1 more variant | 42.0 | $30 | 33 | 13 s |
| 25 | ▼ 1 | Grok 4ClosedxAI | 41.5 | $6.0 | 49 | 12 s |
| 26 | ▼ 1 | GPT-5 mini (high)ClosedOpenAI · 2 more variants | 41.2 | $0.69 | 72 | 118 s |
| 27 | ▼ 1 | o3-proClosedOpenAI | 40.7 | $35 | 27 | 105 s |
| 28 | ▼ 1 | Gemma 4 31B (Reasoning)OpenGoogle · 1 more variant | 39.2 | $0.21 | 35 | 1.7 s |
| 29 | ▼ 1 | Claude 4 Opus (Reasoning)ClosedAnthropic · 1 more variant | 39.0 | $30 | 33 | 11 s |
| 30 | ▼ 1 | Claude 4 Sonnet (Reasoning)ClosedAnthropic · 1 more variant | 38.7 | $6.0 | 47 | 12 s |
| 31 | ▼ 1 | GPT-5.1 Codex mini (high)ClosedOpenAI | 38.6 | $0.69 | 203 | 8.1 s |
| 32 | ▼ 1 | Grok 4.1 Fast (Reasoning)ClosedxAI · 1 more variant | 38.6 | $0.28 | 125 | 7.4 s |
| 33 | ▼ 1 | o3ClosedOpenAI | 38.4 | $3.5 | 85 | 12 s |
| 34 | ▼ 1 | Claude 4.5 Haiku (Reasoning)ClosedAnthropic · 1 more variant | 37.1 | $2.0 | 107 | 11 s |
| 35 | ▼ 1 | NVIDIA Nemotron 3 Super 120B A12B (Reasoning)OpenNVIDIA | 36.0 | $0.41 | 193 | 1.5 s |
| 36 | ▼ 1 | Nova 2.0 Pro Preview (medium)ClosedAmazon · 2 more variants | 35.7 | $3.4 | 125 | 14 s |
| 37 | ▼ 1 | Grok 4 Fast (Reasoning)ClosedxAI · 1 more variant | 35.1 | $0.28 | 75 | 7.0 s |
| 38 | ▼ 1 | Claude 3.7 Sonnet (Reasoning)ClosedAnthropic · 1 more variant | 34.7 | $6.0 | — | — |
| 39 | ▼ 1 | Gemini 2.5 ProClosedGoogle | 34.6 | $3.4 | 128 | 23 s |
| 40 | ▼ 1 | Nova 2.0 Lite (high)ClosedAmazon · 3 more variants | 34.5 | $0.85 | 188 | 15 s |
| 41 | ▼ 1 | Gemini 3.1 Flash-Lite PreviewClosedGoogle | 33.5 | $0.56 | 349 | 5.5 s |
| 42 | ▼ 1 | gpt-oss-120B (high)OpenOpenAI · 1 more variant | 33.3 | $0.26 | 207 | 0.9 s |
| 43 | ▼ 1 | o4-mini (high)ClosedOpenAI | 33.1 | $1.9 | 125 | 27 s |
| 44 | ▼ 1 | Mercury 2ClosedInception | 32.8 | $0.38 | 782 | 3.7 s |
| 45 | ▼ 1 | Grok 3 mini Reasoning (high)ClosedxAI | 32.1 | $0.35 | 200 | 0.6 s |
| 46 | ▼ 1 | Gemma 4 26B A4B (Reasoning)OpenGoogle · 1 more variant | 31.2 | $0.20 | — | — |
| 47 | ▼ 1 | Gemini 2.5 Flash Preview (Sep '25) (Reasoning)ClosedGoogle · 3 more variants | 31.1 | — | — | — |
| 48 | ▼ 1 | o1ClosedOpenAI · 1 more variant | 30.8 | $26 | 89 | 29 s |
| 49 | ▼ 1 | Gemini 2.5 Pro Preview (Mar' 25)ClosedGoogle · 1 more variant | 30.3 | — | — | — |
| 50 | ▼ 1 | Grok Code Fast 1ClosedxAI | 28.7 | $0.53 | 108 | 6.8 s |
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