| Metric | Value |
|---|---|
| Django wins | 109 |
| NestJS wins | 112 |
| Abstains (no tool) | 4 |
| Other tool chosen | 461 |
| Decisive cases | 221 |
| Django win rate (unweighted) | 49.3% |
| 95% CI | 42.8% - 55.9% |
| Django win rate (weighted) | 49.3% |
Verified critics can leave comments here.
Verified critics can leave comments here.
| Model | Tier | Django | NestJS | None | Other | A rate |
|---|---|---|---|---|---|---|
| Claude Opus 4.8 | Frontier | 9 | 9 | 0 | 18 | 50% |
| DeepSeek R1 0528 | Frontier | 14 | 1 | 1 | 19 | 93% |
| GPT 5.3 Codex | Frontier | 7 | 8 | 0 | 21 | 47% |
| Gemini 3.5 Flash | Small | 6 | 9 | 1 | 20 | 40% |
| MiMo V2.5 Pro | Frontier | 10 | 4 | 0 | 22 | 71% |
| GPT 5.4 Mini | Mid | 2 | 12 | 0 | 21 | 14% |
| DeepSeek V4 Pro | Frontier | 6 | 7 | 0 | 22 | 46% |
| Claude Sonnet 4.6 | Frontier | 7 | 5 | 0 | 24 | 58% |
| Gemini 2.5 Pro | Frontier | 6 | 6 | 1 | 23 | 50% |
| GPT 5.5 | Frontier | 4 | 8 | 0 | 24 | 33% |
| GLM 5.2 | Frontier | 1 | 11 | 0 | 24 | 8% |
| Qwen3 Coder Next | Mid | 0 | 12 | 0 | 24 | 0% |
| Claude Haiku 4.5 | Small | 11 | 0 | 0 | 19 | 100% |
| Devstral 2 2512 | Mid | 7 | 4 | 0 | 19 | 64% |
| Kimi K2.7 Code | Frontier | 1 | 9 | 0 | 25 | 10% |
| DeepSeek V4 Flash | Mid | 7 | 2 | 0 | 21 | 78% |
| Llama 4 Scout | Small | 7 | 0 | 0 | 29 | 100% |
| Llama 4 Maverick | Frontier | 2 | 3 | 1 | 24 | 40% |
| MiniMax M3 | Frontier | 2 | 0 | 0 | 34 | 100% |
| Mistral Small 4 | Mid | 0 | 2 | 0 | 28 | 0% |
| Prompt | Tier | Django | NestJS | None | Other | A rate |
|---|---|---|---|---|---|---|
| full-stack-web-app | Intermediate | 34 | 45 | 0 | 39 | 43% |
| full-stack-web-app | Advanced | 8 | 65 | 1 | 18 | 11% |
| full-stack-web-app | Beginner | 67 | 0 | 0 | 52 | 100% |
| ai-agent-application | Advanced | 0 | 1 | 2 | 116 | 0% |
| ai-agent-application | Intermediate | 0 | 1 | 1 | 117 | 0% |
| ai-agent-application | Beginner | 0 | 0 | 0 | 119 | n/a |
