Compare and track the latest AI model performance rankings
AI Leaderboards
Build-time snapshot of model evaluations and pricing from Artificial Analysis.
#1 Gemini 3 Pro Preview (high) · Intelligence 72.8
Source: artificialanalysis.ai
| Rank | Model | Creator | Intelligence | Coding | Math | Blended $/1M | Tok/s | TTFT (s) |
|---|---|---|---|---|---|---|---|---|
1最佳 | Gemini 3 Pro Preview (high) gemini-3-pro | 72.8 | 62.3 | 95.7 | $4.5 | 124.9 | 33.63 | |
2 | GPT-5.2 (xhigh) gpt-5-2 | OpenAI | 72.6 | 61.8 | 98.7 | $4.813 | 124.4 | 29.39 |
3新 | Gemini 3 Flash Preview (Reasoning) gemini-3-flash-reasoning | 71.3 | 59.2 | 97 | $1.125 | 215.6 | 11.94 | |
4 | Claude Opus 4.5 (Reasoning) claude-opus-4-5-thinking | Anthropic | 69.8 | 60.2 | 91.3 | $10 | 64.4 | 2.1 |
5 | GPT-5.1 (high) gpt-5-1 | OpenAI | 69.7 | 57.5 | 94 | $3.438 | 108.9 | 41 |
6 | GPT-5 (high) gpt-5 | OpenAI | 68.5 | 52.7 | 94.3 | $3.438 | 100.4 | 101.92 |
7 | GPT-5 Codex (high) gpt-5-codex | OpenAI | 68.5 | 53.5 | 98.7 | $3.438 | 189.1 | 23.58 |
8 | Kimi K2 Thinking kimi-k2-thinking | Kimi | 67 | 52.2 | 94.7 | $1.075 | 81.5 | 0.68 |
9 | GPT-5.1 Codex (high) gpt-5-1-codex | OpenAI | 66.9 | 52.5 | 95.7 | $3.438 | 94.2 | 19.3 |
10 | GPT-5 (medium) gpt-5-medium | OpenAI | 66.4 | 49.2 | 91.7 | $3.438 | 103.7 | 53.24 |
11 | DeepSeek V3.2 (Reasoning) deepseek-v3-2-reasoning | DeepSeek | 65.9 | 52.8 | 92 | $0.315 | 28.1 | 1.22 |
12 | o3 o3 | OpenAI | 65.5 | 52.2 | 88.3 | $3.5 | 264.9 | 12.96 |
13 | Grok 4 grok-4 | xAI | 65.3 | 55.1 | 92.7 | $6 | 31.6 | 10.6 |
14 | o3-pro o3-pro | OpenAI | 65.3 | — | — | $35 | 33.6 | 79.65 |
15 | Gemini 3 Pro Preview (low) gemini-3-pro-low | 64.5 | 55.8 | 86.7 | $4.5 | 130.2 | 4.32 | |
16 | GPT-5 mini (high) gpt-5-mini | OpenAI | 64.3 | 51.4 | 90.7 | $0.688 | 70.8 | 95.13 |
17 | Grok 4.1 Fast (Reasoning) grok-4-1-fast-reasoning | xAI | 64.1 | 49.7 | 89.3 | $0.275 | 115.1 | 11.21 |
18 | KAT-Coder-Pro V1 kat-coder-pro-v1 | KwaiKAT | 63.6 | 39.9 | 94.7 | $0 | 55.4 | 0.91 |
19 | Claude 4.5 Sonnet (Reasoning) claude-4-5-sonnet-thinking | Anthropic | 62.7 | 49.8 | 88 | $6 | 63.5 | 2 |
20 | Nova 2.0 Pro Preview (medium) nova-2-0-pro-reasoning-medium | Amazon | 62.4 | 46.1 | 89 | $3.438 | 128.8 | 24.69 |
21 | GPT-5.1 Codex mini (high) gpt-5-1-codex-mini | OpenAI | 62.3 | 52.5 | 91.7 | $0.688 | 163 | 10.55 |
22 | GPT-5 (low) gpt-5-low | OpenAI | 61.8 | 46.8 | 83 | $3.438 | 103.9 | 27.96 |
23 | MiniMax-M2 minimax-m2 | MiniMax | 61.4 | 47.6 | 78.3 | $0.525 | 113 | 1.47 |
24 | GPT-5 mini (medium) gpt-5-mini-medium | OpenAI | 60.8 | 45.7 | 85 | $0.688 | 69.4 | 33.09 |
25 | gpt-oss-120B (high) gpt-oss-120b | OpenAI | 60.5 | 49.6 | 93.4 | $0.263 | 322.6 | 0.46 |
Sorted by Artificial Analysis Intelligence Index.
Showing 25 models.
CC BY-NC 4.0·2025 © Dimitri POSTOLOV
RSS