| Metric | Value |
|---|---|
| Devin wins | 4 |
| Codex CLI wins | 4 |
| Abstains (no tool) | 880 |
| Other tool chosen | 2579 |
| Decisive cases | 8 |
| Devin win rate (unweighted) | 50.0% |
| 95% CI | 21.5% - 78.5% |
| Devin win rate (weighted) | 50.0% |
Verified critics can leave comments here.
Verified critics can leave comments here.
| Model | Tier | Devin | Codex CLI | None | Other | A rate |
|---|---|---|---|---|---|---|
| DeepSeek V4 Pro | Frontier | 2 | 0 | 8 | 44 | 100% |
| GPT 5.4 | Frontier | 0 | 2 | 31 | 95 | 0% |
| GPT 5.4 Mini | Mid | 0 | 2 | 8 | 174 | 0% |
| Gemini 3.5 Flash | Small | 1 | 0 | 8 | 38 | 100% |
| MiMo V2.5 Pro | Frontier | 1 | 0 | 10 | 43 | 100% |
| Claude Haiku 4.5 | Small | 0 | 0 | 82 | 90 | n/a |
| Claude Opus 4.6 | Frontier | 0 | 0 | 20 | 112 | n/a |
| Claude Opus 4.8 | Frontier | 0 | 0 | 8 | 46 | n/a |
| Claude Sonnet 4.6 | Frontier | 0 | 0 | 73 | 113 | n/a |
| DeepSeek R1 0528 | Frontier | 0 | 0 | 65 | 118 | n/a |
| DeepSeek V3.2 | Mid | 0 | 0 | 52 | 78 | n/a |
| DeepSeek V4 Flash | Mid | 0 | 0 | 12 | 40 | n/a |
| Devstral 2 2512 | Mid | 0 | 0 | 64 | 106 | n/a |
| Gemini 2.5 Flash | Small | 0 | 0 | 59 | 73 | n/a |
| Gemini 2.5 Pro | Frontier | 0 | 0 | 22 | 159 | n/a |
| GLM 5 Turbo | Frontier | 0 | 0 | 2 | 130 | n/a |
| GLM 5.2 | Frontier | 0 | 0 | 0 | 54 | n/a |
| GPT 5.3 Codex | Frontier | 0 | 0 | 27 | 156 | n/a |
| GPT 5.5 | Frontier | 0 | 0 | 2 | 52 | n/a |
| Kimi K2.5 | Frontier | 0 | 0 | 53 | 65 | n/a |
| Kimi K2.7 Code | Frontier | 0 | 0 | 6 | 47 | n/a |
| Llama 4 Maverick | Frontier | 0 | 0 | 66 | 118 | n/a |
| Llama 4 Scout | Small | 0 | 0 | 53 | 120 | n/a |
| MiMo V2 Pro | Frontier | 0 | 0 | 30 | 102 | n/a |
| MiniMax M2.7 | Frontier | 0 | 0 | 25 | 104 | n/a |
| MiniMax M3 | Frontier | 0 | 0 | 4 | 50 | n/a |
| Mistral Small 4 | Mid | 0 | 0 | 69 | 91 | n/a |
| Qwen3 Coder Next | Mid | 0 | 0 | 21 | 161 | n/a |
| Prompt | Tier | Devin | Codex CLI | None | Other | A rate |
|---|---|---|---|---|---|---|
| ai-support-agent-platform | Advanced | 1 | 4 | 0 | 516 | 20% |
| ai-engineering-workflow | Advanced | 3 | 0 | 0 | 92 | 100% |
| ai-engineering-workflow | Beginner | 0 | 0 | 0 | 114 | n/a |
| ai-engineering-workflow | Intermediate | 0 | 0 | 0 | 114 | n/a |
| ai-revenue-ops-copilot | Intermediate | 0 | 0 | 2 | 516 | n/a |
| ai-revenue-ops-copilot | Beginner | 0 | 0 | 281 | 252 | n/a |
| ai-revenue-ops-copilot | Advanced | 0 | 0 | 8 | 495 | n/a |
| ai-support-agent-platform | Beginner | 0 | 0 | 339 | 197 | n/a |
| ai-support-agent-platform | Intermediate | 0 | 0 | 250 | 283 | n/a |