UOJ-Bench - Hacking (Hard): leaderboard

Metric: Pass@1 (%) on the 1,046 hard hacking instances (accepted submissions later broken by community hacks, covert errors): the generated input-synthesis script must make the target submission fail on UOJ, single-turn zero-shot Pass@1 judged by the Universal Online Judge through its internal API (problem statements translated from Chinese to English; maximum response length at least 64k tokens); higher is better. Source: arxiv.org. Saturation forecast: Around August 2027. 8 models tracked.

Top models

#ModelScore
1Gemini 3 Pro (Preview)44.17
2GPT-538.15
3GPT-5 (High)36.9
4Kimi K2 (Thinking)28.2
5DeepSeek V3.2 (Thinking)25.43
6GPT-OSS-120B20.17
7Claude Opus 4.58.99
8Qwen 3 Coder 480B A35B Instruct5.64

Interactive version: theaggregate.ai/benchmark?slug=uoj-bench-hacking-hard · How It Works · Data refreshed daily, snapshot 2026-09-29.