Wikibench

OJBench

reasoning

OJBench is a competition-level code benchmark designed to assess the competitive-level code reasoning abilities of large language models. It comprises 232 programming competition problems from NOI and ICPC, categorized into Easy, Medium, and Hard difficulty levels. The benchmark evaluates models' ability to solve complex competitive programming challenges using Python and C++.

Leaderboard

Showing 9 of 9 results

  1. Kimi K2.6
    60.6%
    i
  2. Kimi K2-Thinking-0905
    48.7%
    i
  3. Qwen3.5-27B
    40.1%
    i
  4. Qwen3.5-122B-A10B
    39.5%
    i
  5. Qwen3.5-35B-A3B
    36.0%
    i
  6. Qwen3-235B-A22B-Thinking-2507
    32.5%
    i
  7. Qwen3-Next-80B-A3B-Thinking
    29.7%
    i
  8. Kimi K2 Instruct
    27.1%
    i
  9. Kimi K2-Instruct-0905
    27.1%
    i