Multi-LCB - Pass@5 T0.6 - C#: leaderboard

Metric: Pass@5 (%). Source: arxiv.org. 25 models tracked.

Top models

#ModelScore
1GPT-OSS-120B (Medium)80.7
2GPT-OSS-20B (Medium)75.5
3Qwen 3 235B A22B (Thinking)67.1
4Qwen 3 32B (Thinking)66
5GPT-OSS-120B (Low)63.4
6Qwen 3 14B (Reasoning)60.4
7Qwen 3 30B A3B (Thinking)58.1
8GPT-OSS-20B (Low)54.6
9Qwen 3 8B (Thinking)51.8

Interactive version: theaggregate.ai/benchmark?slug=multi-lcb-pass-5-t0-6-c · How It Works · Data refreshed daily, snapshot 2026-09-19.