| Metric | Value |
|---|---|
| TypeScript wins | 238 |
| Python wins | 693 |
| Abstains (no tool) | 26 |
| Other tool chosen | 109 |
| Decisive cases | 931 |
| TypeScript win rate (unweighted) | 25.6% |
| 95% CI | 22.9% - 28.5% |
| TypeScript win rate (weighted) | 25.6% |
Verified critics can leave comments here.
Verified critics can leave comments here.
| Model range | Tier | TypeScript | Python | None | Other | A rate |
|---|---|---|---|---|---|---|
| DeepSeek Flash 3.2–4.1 | Mid | 9 | 35 | 0 | 9 | 20% |
| DeepSeek Pro 4–4 0813 | Frontier | 15 | 34 | 3 | 2 | 31% |
| DeepSeek R1 | Frontier | 2 | 41 | 5 | 6 | 5% |
| Devstral | Mid | 8 | 34 | 0 | 0 | 19% |
| Gemini Flash 2.5–3.8 | Small | 18 | 35 | 0 | 1 | 34% |
| Gemini Pro 2.5–3.1 Preview | Frontier | 16 | 33 | 1 | 4 | 33% |
| GLM 5 Turbo–5.3 | Frontier | 22 | 30 | 0 | 2 | 42% |
| GPT 5.4–6 Astra | Frontier | 14 | 37 | 0 | 3 | 27% |
| GPT Coding 5.3 Codex–6.1 Sol | Frontier | 11 | 42 | 1 | 0 | 21% |
| GPT Mini 5.4 Mini–6 Luna | Mid | 19 | 32 | 0 | 3 | 37% |
| Haiku | Small | 1 | 44 | 0 | 9 | 2% |
| Kimi 2.5–3 | Frontier | 22 | 26 | 4 | 2 | 46% |
| Llama Maverick | Frontier | 6 | 26 | 6 | 16 | 19% |
| Llama Scout | Small | 0 | 42 | 0 | 12 | 0% |
| MiMo Pro 2–2.6 | Frontier | 9 | 43 | 0 | 2 | 17% |
| MiniMax M 2.7–3 | Frontier | 21 | 29 | 0 | 4 | 42% |
| Mistral Small | Mid | 1 | 26 | 0 | 26 | 4% |
| Opus 4.6–5.5 | Frontier | 15 | 39 | 0 | 0 | 28% |
| Qwen Coder | Mid | 16 | 25 | 6 | 7 | 39% |
| Sonnet 4.6–5.5 | Frontier | 13 | 40 | 0 | 1 | 25% |
| Prompt | Tier | TypeScript | Python | None | Other | A rate |
|---|---|---|---|---|---|---|
| ai-agent-application | Beginner | 1 | 174 | 2 | 1 | 1% |
| ai-agent-application | Intermediate | 1 | 170 | 7 | 0 | 1% |
| ai-agent-application | Advanced | 1 | 163 | 12 | 0 | 1% |
| full-stack-web-app | Intermediate | 94 | 55 | 0 | 29 | 63% |
| full-stack-web-app | Beginner | 39 | 102 | 4 | 33 | 28% |
| full-stack-web-app | Advanced | 102 | 29 | 1 | 46 | 78% |
