Coding leaderboard from coding-specific benchmark metrics.
Domain score averages relative percentile across included metrics.
Domain score 100
Coding leaderboard from coding-specific benchmark metrics.
| Rank | Model | Creator | Domain Score | Speed | Blended Price |
|---|---|---|---|---|---|
| #1 | GPT-5.5 (xhigh) | OpenAI | 99.7 | 62.4 tok/s | $11.25/M |
| #2 |
| OpenAI |
| 99.7 |
| 75.5 tok/s |
| $5.63/M |
| #3 | GPT-5.5 (high) | OpenAI | 99.5 | 62.3 tok/s | $11.25/M |
| #4 | Gemini 3.1 Pro Preview | 99.4 | 135.9 tok/s | $4.50/M |
| #5 | Claude Opus 4.8 (Adaptive Reasoning, Max Effort) | Anthropic | 98.9 | 66.3 tok/s | $10.94/M |
| #6 | GPT-5.5 (medium) | OpenAI | 98.7 | 59.2 tok/s | $11.25/M |
| #7 | Claude Opus 4.7 (Adaptive Reasoning, Max Effort) | Anthropic | 98.5 | 61.8 tok/s | $10.94/M |
| #8 | GPT-5.3 Codex (xhigh) | OpenAI | 98.2 | 98.6 tok/s | $4.81/M |
| #9 | Gemini 3 Pro Preview (high) | 98.1 | n/a | $4.50/M |
| #10 | GPT-5.2 (xhigh) | OpenAI | 97.6 | 71 tok/s | $4.81/M |
| #11 | GPT-5.5 (low) | OpenAI | 97.4 | 60.2 tok/s | $11.25/M |
| #12 | Claude Opus 4.7 (Non-reasoning, High Effort) | Anthropic | 97.2 | 48.1 tok/s | $10.94/M |
| #13 | Claude Opus 4.6 (Adaptive Reasoning, Max Effort) | Anthropic | 96.8 | 47.3 tok/s | $10.94/M |
| #14 | Kimi K2.6 | Kimi | 96.7 | 41.6 tok/s | $1.71/M |
| #15 | Claude Opus 4.5 (Reasoning) | Anthropic | 96.3 | 53.5 tok/s | $10.94/M |
| #16 | GPT-5.4 mini (xhigh) | OpenAI | 96.3 | 176 tok/s | $1.69/M |
| #17 | Muse Spark | Meta | 96.2 | n/a | - |
| #18 | DeepSeek V4 Pro (Reasoning, Max Effort) | DeepSeek | 95.7 | 61.8 tok/s | $0.544/M |
| #19 | Qwen3.7 Max | Alibaba | 95.7 | 186.5 tok/s | $3.75/M |
| #20 | Gemini 3.5 Flash (high) | 95.4 | 212.4 tok/s | $3.38/M |
| #21 | Gemini 3 Flash Preview (Reasoning) | 95.4 | 172.8 tok/s | $1.13/M |
| #22 | GPT-5.5 (Non-reasoning) | OpenAI | 95.3 | 59 tok/s | $11.25/M |
| #23 | GPT-5.4 (low) | OpenAI | 95.2 | 63.6 tok/s | $5.63/M |
| #24 | Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) | Anthropic | 94.9 | 70.4 tok/s | $6.56/M |
| #25 | Gemini 3.5 Flash (minimal) | 94.9 | 199.1 tok/s | $3.38/M |
| #26 | GPT-5.2 Codex (xhigh) | OpenAI | 94.7 | 105.3 tok/s | $4.81/M |
| #27 | MiMo-V2.5-Pro | Xiaomi | 94.7 | 43.3 tok/s | $0.544/M |
| #28 | Gemini 3.5 Flash (medium) | 94.7 | 210.2 tok/s | $3.38/M |
| #29 | GPT-5.5 Instant (May 2026) | OpenAI | 94.7 | n/a | $11.25/M |
| #30 | GPT-5.2 (medium) | OpenAI | 94.2 | n/a | $4.81/M |
| #31 | Claude Opus 4.6 (Non-reasoning, High Effort) | Anthropic | 93.7 | 40.9 tok/s | $10.94/M |
| #32 | Claude Sonnet 4.6 (Non-reasoning, High Effort) | Anthropic | 93.6 | 57.9 tok/s | $6.56/M |
| #33 | Gemini 3 Pro Preview (low) | 92.9 | n/a | $4.50/M |
| #34 | Qwen3.6 Max Preview | Alibaba | 92.7 | 40.9 tok/s | $2.93/M |
| #35 | Qwen3.7 Plus | Alibaba | 92.6 | 53.6 tok/s | $0.590/M |
| #36 | GPT-5.1 (high) | OpenAI | 92.5 | 121.2 tok/s | $3.44/M |
| #37 | GPT-5.4 nano (xhigh) | OpenAI | 92.2 | 158.3 tok/s | $0.463/M |
| #38 | GLM-5 (Reasoning) | Z AI | 92.0 | 79.5 tok/s | $1.55/M |
| #39 | DeepSeek V4 Pro (Reasoning, High Effort) | DeepSeek | 91.4 | 57.6 tok/s | $0.544/M |
| #40 | MiniMax-M2.7 | MiniMax | 91.1 | 75 tok/s | $0.525/M |
| #41 | GPT-5.4 (Non-reasoning) | OpenAI | 91.0 | 59.3 tok/s | $5.63/M |
| #42 | Grok 4.3 (high) | xAI | 91.0 | 237.5 tok/s | $1.56/M |
| #43 | Kimi K2.5 (Reasoning) | Kimi | 90.9 | 31.7 tok/s | $1.19/M |
| #44 | Grok 4 | xAI | 90.9 | n/a | $11.00/M |
| #45 | DeepSeek V3.2 Speciale | DeepSeek | 90.9 | n/a | - |
| #46 | MiniMax-M3 | MiniMax | 90.8 | 45.6 tok/s | $0.525/M |
| #47 | Gemini 3 Flash Preview (Non-reasoning) | 90.5 | 181.3 tok/s | $1.13/M |
| #48 | GLM-4.7 (Reasoning) | Z AI | 90.2 | 79.2 tok/s | $1.00/M |
| #49 | GLM-5.1 (Reasoning) | Z AI | 89.8 | 46.8 tok/s | $2.15/M |
| #50 | Claude Sonnet 4.6 (Non-reasoning, Low Effort) | Anthropic | 89.7 | 61.7 tok/s | $6.56/M |
| #51 | Claude Opus 4.5 (Non-reasoning) | Anthropic | 89.7 | 47.6 tok/s | $10.94/M |
| #52 | Grok 4.20 0309 (Reasoning) | xAI | 89.6 | 166.5 tok/s | $3.00/M |
| #53 | Grok 4.20 0309 v2 (Reasoning) | xAI | 89.3 | 168.7 tok/s | $3.00/M |
| #54 | MiMo-V2.5 | Xiaomi | 88.2 | 77.4 tok/s | $0.175/M |
| #55 | DeepSeek V4 Flash (Reasoning, Max Effort) | DeepSeek | 88.1 | 107.8 tok/s | $0.175/M |
| #56 | GPT-5 (high) | OpenAI | 87.7 | 111.1 tok/s | $3.44/M |
| #57 | GPT-5 Codex (high) | OpenAI | 87.7 | 171.1 tok/s | $3.44/M |
| #58 | GPT-5.1 Codex mini (high) | OpenAI | 87.3 | 213.6 tok/s | $0.688/M |
| #59 | MiMo-V2-Pro | Xiaomi | 87.2 | 42.5 tok/s | $1.50/M |
| #60 | Gemini 2.5 Pro Preview (Mar' 25) | 86.9 | n/a | - |
| #61 | Gemma 4 31B (Reasoning) | 86.8 | 34.8 tok/s | - |
| #62 | GPT-5.4 mini (medium) | OpenAI | 86.8 | 184.4 tok/s | $1.69/M |
| #63 | Kimi K2 Thinking | Kimi | 86.6 | 131.1 tok/s | $1.08/M |
| #64 | o3 | OpenAI | 86.5 | 159.6 tok/s | $3.50/M |
| #65 | Qwen3.5 397B A17B (Reasoning) | Alibaba | 86.3 | 51.8 tok/s | $1.35/M |
| #66 | Gemini 2.5 Pro Preview (May' 25) | 86.0 | n/a | $3.44/M |
| #67 | DeepSeek V4 Flash (Reasoning, High Effort) | DeepSeek | 86.0 | n/a | $0.175/M |
| #68 | Claude 4.5 Sonnet (Reasoning) | Anthropic | 85.9 | 50.1 tok/s | $6.56/M |
| #69 | GPT-5.1 Codex (high) | OpenAI | 85.7 | 182.1 tok/s | $3.44/M |
| #70 | GLM-5-Turbo | Z AI | 85.6 | n/a | - |
| #71 | Qwen3.6 Plus | Alibaba | 85.4 | 52.8 tok/s | $1.13/M |
| #72 | DeepSeek V4 Pro (Non-reasoning) | DeepSeek | 85.3 | 61.2 tok/s | $0.544/M |
| #73 | Grok 4.1 Fast (Reasoning) | xAI | 84.8 | n/a | - |
| #74 | MiniMax-M2.5 | MiniMax | 84.8 | 202.9 tok/s | $0.525/M |
| #75 | GLM 5V Turbo (Reasoning) | Z AI | 84.4 | n/a | - |
| #76 | DeepSeek V3.2 (Reasoning) | DeepSeek | 84.3 | n/a | $0.337/M |
| #77 | Grok 4.3 (medium) | xAI | 84.3 | 197.6 tok/s | $1.56/M |
| #78 | Gemini 2.5 Pro | 84.1 | 139.8 tok/s | $3.44/M |
| #79 | o4-mini (high) | OpenAI | 84.0 | 151 tok/s | $1.93/M |
| #80 | GPT-5 (medium) | OpenAI | 83.3 | 85.6 tok/s | $3.44/M |
