CAPA: leaderboard
Metric: Executable success (%; share of the 300 held-out coding sessions solved within an eight-turn budget, a submission counting when the execution-grounded judge accepts it from the hidden tests and the reference solution, and a simulated user agent answering clarification questions; same-user history: the model also sees the dialogue traces of five resolved sessions from the same user, whose recurring ambiguity pattern the held-out request repeats). Source: arxiv.org. Saturation forecast: Around December 2026. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.8 | 90 |
| 2 | GLM-5.2 | 89.7 |
| 3 | GPT-5.5 | 84.3 |
| 4 | Claude Sonnet 4.6 | 83.7 |
| 5 | Qwen 3.7 Max | 82.7 |
| 6 | Kimi K2.6 | 82 |
| 7 | Gemini 3.5 Flash | 81.7 |
| 8 | DeepSeek V4 Pro | 79 |
| 9 | GPT-5.6 Sol | 78.7 |
| 10 | Qwen 3.5 27B | 74.3 |
| 11 | Llama 3.3 70B Instruct | 30 |
| 12 | Qwen 3 8B | 27.7 |
Interactive version: theaggregate.ai/benchmark?slug=capa · How It Works · Data refreshed daily, snapshot 2026-09-29.