Sphinx - C++: leaderboard
Sphinx checklist-coverage score on its 500 C++ pull requests, 450 needing changes and 50 bug-free controls whose correct review is no comment: a GPT-4o judge checks each generated review against a checklist derived from the merged change, buggy cases weighted 0.9; percent, higher is better. Microsoft and University of Rochester, 2026.
Metric: Checklist Coverage Score (%; GPT-4o judge, 450 buggy + 50 bug-free PRs per language). Source: arxiv.org. Saturation forecast: Estimated already saturated. 29 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4.1 | 41.85 | #492 |
| 2 | Gemini 2.5 Pro | 38.59 | #262 |
| 3 | O3 Mini | 37.68 | #536 |
| 4 | Claude 3.7 Sonnet | 35.68 | #455 |
| 5 | O4 Mini | 34.93 | #354 |
| 6 | GPT-4o (2024-08-06) | 34.39 | #731 |
| 7 | Qwen 3 30B A3B | 34.02 | #1124 |
| 8 | Claude 3.5 Sonnet | 33.63 | #747 |
| 9 | Qwen 2.5 72B Instruct | 33.1 | #1093 |
| 10 | GPT-4o Mini | 32.8 | #1627 |
| 11 | Qwen 3 235B A22B | 32.51 | #622 |
| 12 | GPT-4.1 Nano | 31.48 | #1992 |
| 13 | GPT-4.1 Mini | 31.38 | #775 |
| 14 | Qwen 3 32B | 29.99 | #960 |
| 15 | DeepSeek V3 | 28.58 | #705 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=sphinx-c-plus-plus · How It Works · Data refreshed daily, snapshot 2026-11-12.