| Metric | Value |
|---|---|
| Claude Code wins | 48 |
| OpenAI wins | 50 |
| Abstains (no tool) | 875 |
| Other tool chosen | 2547 |
| Decisive cases | 98 |
| Claude Code win rate (unweighted) | 49.0% |
| 95% CI | 39.3% - 58.7% |
| Claude Code win rate (weighted) | 49.0% |
Verified critics can leave comments here.
Verified critics can leave comments here.
| Model | Tier | Claude Code | OpenAI | None | Other | A rate |
|---|---|---|---|---|---|---|
| Qwen3 Coder Next | Mid | 0 | 41 | 21 | 124 | 0% |
| MiniMax M2.7 | Frontier | 9 | 0 | 23 | 91 | 100% |
| GPT 5.4 Mini | Mid | 8 | 0 | 7 | 172 | 100% |
| Claude Opus 4.8 | Frontier | 7 | 0 | 9 | 47 | 100% |
| GPT 5.5 | Frontier | 7 | 0 | 3 | 53 | 100% |
| MiniMax M3 | Frontier | 6 | 0 | 4 | 53 | 100% |
| Llama 4 Maverick | Frontier | 0 | 5 | 66 | 116 | 0% |
| GPT 5.4 | Frontier | 3 | 1 | 30 | 88 | 75% |
| DeepSeek V3.2 | Mid | 2 | 0 | 50 | 72 | 100% |
| DeepSeek V4 Pro | Frontier | 2 | 0 | 11 | 50 | 100% |
| Mistral Small 4 | Mid | 2 | 0 | 69 | 91 | 100% |
| Claude Haiku 4.5 | Small | 0 | 2 | 82 | 91 | 0% |
| DeepSeek V4 Flash | Mid | 1 | 0 | 14 | 45 | 100% |
| GLM 5.2 | Frontier | 1 | 0 | 0 | 62 | 100% |
| Devstral 2 2512 | Mid | 0 | 1 | 63 | 100 | 0% |
| Claude Opus 4.6 | Frontier | 0 | 0 | 19 | 107 | n/a |
| Claude Sonnet 4.6 | Frontier | 0 | 0 | 74 | 115 | n/a |
| DeepSeek R1 0528 | Frontier | 0 | 0 | 66 | 119 | n/a |
| Gemini 2.5 Flash | Small | 0 | 0 | 56 | 70 | n/a |
| Gemini 2.5 Pro | Frontier | 0 | 0 | 21 | 163 | n/a |
| Gemini 3.5 Flash | Small | 0 | 0 | 10 | 46 | n/a |
| GLM 5 Turbo | Frontier | 0 | 0 | 2 | 124 | n/a |
| GPT 5.3 Codex | Frontier | 0 | 0 | 26 | 159 | n/a |
| Kimi K2.5 | Frontier | 0 | 0 | 50 | 63 | n/a |
| Kimi K2.7 Code | Frontier | 0 | 0 | 6 | 56 | n/a |
| Llama 4 Scout | Small | 0 | 0 | 51 | 123 | n/a |
| MiMo V2 Pro | Frontier | 0 | 0 | 29 | 97 | n/a |
| MiMo V2.5 Pro | Frontier | 0 | 0 | 13 | 50 | n/a |
| Prompt | Tier | Claude Code | OpenAI | None | Other | A rate |
|---|---|---|---|---|---|---|
| ai-revenue-ops-copilot | Advanced | 14 | 22 | 7 | 461 | 39% |
| ai-support-agent-platform | Advanced | 15 | 4 | 1 | 501 | 79% |
| ai-revenue-ops-copilot | Intermediate | 9 | 9 | 2 | 498 | 50% |
| ai-support-agent-platform | Beginner | 3 | 5 | 336 | 192 | 38% |
| ai-revenue-ops-copilot | Beginner | 1 | 6 | 280 | 246 | 14% |
| ai-engineering-workflow | Advanced | 5 | 0 | 0 | 105 | 100% |
| ai-support-agent-platform | Intermediate | 0 | 4 | 249 | 280 | 0% |
| ai-engineering-workflow | Intermediate | 1 | 0 | 0 | 131 | 100% |
| ai-engineering-workflow | Beginner | 0 | 0 | 0 | 133 | n/a |