| Metric | Value |
|---|---|
| Go wins | 31 |
| TypeScript wins | 215 |
| Abstains (no tool) | 25 |
| Other tool chosen | 633 |
| Decisive cases | 246 |
| Go win rate (unweighted) | 12.6% |
| 95% CI | 9.0% - 17.3% |
| Go win rate (weighted) | 12.6% |
Verified critics can leave comments here.
Verified critics can leave comments here.
| Model range | Tier | Go | TypeScript | None | Other | A rate |
|---|---|---|---|---|---|---|
| DeepSeek Flash 3.2–4.1 | Mid | 8 | 6 | 0 | 32 | 57% |
| DeepSeek Pro 4–4 0813 | Frontier | 1 | 13 | 3 | 31 | 7% |
| DeepSeek R1 | Frontier | 2 | 2 | 5 | 38 | 50% |
| Devstral | Mid | 0 | 8 | 0 | 28 | 0% |
| Gemini Flash 2.5–3.8 | Small | 1 | 16 | 0 | 31 | 6% |
| Gemini Pro 2.5–3.1 Preview | Frontier | 0 | 13 | 1 | 34 | 0% |
| GLM 5 Turbo–5.3 | Frontier | 0 | 22 | 0 | 26 | 0% |
| GPT 5.4–6 Astra | Frontier | 0 | 14 | 0 | 34 | 0% |
| GPT Coding 5.3 Codex–6.1 Sol | Frontier | 0 | 11 | 1 | 36 | 0% |
| GPT Mini 5.4 Mini–6 Luna | Mid | 2 | 19 | 0 | 27 | 10% |
| Haiku | Small | 8 | 1 | 0 | 39 | 89% |
| Kimi 2.5–3 | Frontier | 1 | 20 | 4 | 23 | 5% |
| Llama Maverick | Frontier | 0 | 5 | 6 | 32 | 0% |
| Llama Scout | Small | 0 | 0 | 0 | 39 | n/a |
| MiMo Pro 2–2.6 | Frontier | 0 | 7 | 0 | 39 | 0% |
| MiniMax M 2.7–3 | Frontier | 0 | 19 | 0 | 26 | 0% |
| Mistral Small | Mid | 8 | 1 | 0 | 24 | 89% |
| Opus 4.6–5.5 | Frontier | 0 | 13 | 0 | 35 | 0% |
| Qwen Coder | Mid | 0 | 13 | 5 | 23 | 0% |
| Sonnet 4.6–5.5 | Frontier | 0 | 12 | 0 | 36 | 0% |
| Prompt | Tier | Go | TypeScript | None | Other | A rate |
|---|---|---|---|---|---|---|
| full-stack-web-app | Advanced | 29 | 92 | 1 | 34 | 24% |
| full-stack-web-app | Intermediate | 2 | 84 | 0 | 51 | 2% |
| full-stack-web-app | Beginner | 0 | 36 | 4 | 99 | 0% |
| ai-agent-application | Advanced | 0 | 1 | 11 | 144 | 0% |
| ai-agent-application | Beginner | 0 | 1 | 2 | 155 | 0% |
| ai-agent-application | Intermediate | 0 | 1 | 7 | 150 | 0% |
