PBT-Bench (Baseline Prompt): leaderboard
Metric: Bug recall (%) on the 100 problems under the open-ended baseline prompt (write tests that expose the injected bugs, no property-based testing scaffolding), OpenHands agent scaffold, mean of three runs, each problem weighted equally; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 8 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 4.6 | 76.7 |
| 2 | GLM-5.1 | 66.6 |
| 3 | DeepSeek V3.2 | 64.2 |
| 4 | Gemini 3 Flash | 56.2 |
| 5 | Qwen 3.6 Plus | 53.5 |
| 6 | Grok 4.1 Fast | 50.1 |
| 7 | Step 3.5 Flash | 37.8 |
Interactive version: theaggregate.ai/benchmark?slug=pbt-bench-baseline-prompt · How It Works · Data refreshed daily, snapshot 2026-10-07.