SuperCLUE-Project - Overall: leaderboard
SuperCLUE-Project evaluates project-level code generation: building multi-file, structured software projects. Headline board of the newest workbook the superclueai.com vertical lists; score 0-100, higher is better.
Metric: Score. Source: www.superclueai.com. Saturation forecast: Estimated already saturated. 22 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | O3 Mini (High) | 82.08 | #282 (O3 Mini) |
| 2 | Gemini 2.5 Pro (Preview 03-25) | 81.75 | #126 |
| 3 | Claude 3.7 Sonnet (Thinking) | 81.63 | #259 (Claude 3.7 Sonnet) |
| 4 | O4 Mini (High) | 81.35 | #185 (O4 Mini) |
| 5 | GPT-4.1 | 81.2 | #258 |
| 6 | DeepSeek V3 (0324) | 79.43 | #349 |
| 7 | DeepSeek R1 | 78.59 | #262 |
| 8 | Qwen 3 235B A22B | 78.04 | #323 |
| 9 | Claude 3.5 Sonnet | 77.94 | #362 |
| 10 | Qwen 3 32B | 77.37 | #449 |
| 11 | QwQ-32B | 76.55 | #422 |
| 12 | Qwen 3 30B A3B | 74.86 | #496 |
| 13 | Llama 4 Maverick Instruct FP8 | 73.8 | #424 |
| 14 | GLM-Zero-Preview | 71.2 | #682 |
| 15 | Llama 4 Scout Instruct | 67.45 | #547 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=superclue-project-overall · How It Works · Data refreshed daily, snapshot 2026-11-05.