Multi-LCB - Pass@5 T0.6 - Java: leaderboard
Metric: Pass@5 (%). Source: arxiv.org. 25 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-OSS-120B (Medium) | 84.8 |
| 2 | GPT-OSS-20B (Medium) | 80.1 |
| 3 | Qwen 3 235B A22B (Thinking) | 70.7 |
| 4 | GPT-OSS-120B (Low) | 70.5 |
| 5 | Qwen 3 32B (Thinking) | 69.5 |
| 6 | Qwen 3 30B A3B (Thinking) | 64.8 |
| 7 | Qwen 3 14B (Reasoning) | 63.6 |
| 8 | GPT-OSS-20B (Low) | 59.1 |
| 9 | Qwen 3 8B (Thinking) | 57.4 |
Interactive version: theaggregate.ai/benchmark?slug=multi-lcb-pass-5-t0-6-java · How It Works · Data refreshed daily, snapshot 2026-09-19.