Agents Directory
SkillsRankingsAgents
CategoriesModelsBenchmarksCompareAgent LeaderboardSkillsRankingsAgentsAbout
/Benchmarks
/LiveCodeBench v6
L

LiveCodeBench v6

Coding

Contamination-free competitive programming: problems are continuously collected from LeetCode, AtCoder and Codeforces after model cutoffs and scored as pass@1. Higher is better.

Official source
The official LiveCodeBench site has not been refreshed past the OpenAIo3/Gemini 2.5 era. Rows here from aggregators include vendor-reported 2026 scores, but ClaudeClaude Fable 5, ClaudeClaude Opus 4.8 and OpenAIGPT-5.5 are not on the board yet.

LiveCodeBench continuously scrapes new problems from LeetCode, AtCoder and Codeforces contests, tagging each with its publication date so models can be evaluated only on problems released after their training cutoff. Release v6 contains 1,055 problems spanning May 2023 through April 2025. The headline metric is pass@1 on code generation, computed with test-case checkers, and the official leaderboard exposes a date-range slider that recomputes scores for the selected window while flagging potentially contaminated models. Most 2026 frontier scores on this board are vendor-reported on the full v6 set via aggregators, since the official site's own table has not been refreshed past the o3/Gemini 2.5 era.

Leaderboard
#ModelScoreProvider
  • 1
    DeepSeekDeepSeek V4Pro Max
    93.5%DeepSeek
  • 2
    GeminiGemini 3 ProHigh
    91.7%Google DeepMind
  • 3
    QwenQwen3.7 Max
    91.6%Qwen
  • 4
    DeepSeekDeepSeek V4 FlashMax
    91.6%DeepSeek
  • 5
    GeminiGemini 3 FlashReasoning
    90.8%Google DeepMind
  • 6
    MoonshotAIKimi K2.6
    89.6%Moonshot AI
  • 7
    QwenQwen3.6 Plus
    87.1%Qwen
  • 8
    StepfunStep 3.5 Flash
    86.4%StepFun
  • 9
    MoonshotAIKimi K2.5
    85%Moonshot AI
  • 10
    GLM 4.7 logoGLM 4.7
    84.9%Z.AI
  • 11
    ClaudeClaude Opus 4.5
    84.8%Anthropic
  • 12
    MoonshotAIKimi K2 Thinking
    83.1%Moonshot AI
  • 13
    GLM 4.6 logoGLM 4.6
    82.8%Z.AI
  • 14
    SSeed-2.0-Lite
    81.7%ByteDance Seed
  • 15
    OpenAIo4 Mini High
    80.2%OpenAI
  • 16
    OpenAIo3High
    75.8%OpenAI
  • 17
    GeminiGemini 2.5 Pro Preview 06-05
    73.6%Google DeepMind
Sources:
Official results JSON (performances_generation.json)LiveCodeBench official leaderboardLiveCodeBench paper (arXiv 2403.07974)LiveCodeBench/LiveCodeBench
Share:
Details:

  • Category

    Coding

  • LCreated by

    LiveCodeBench

  • Models tested

    17

  • Leader

    DeepSeekDeepSeek V4

  • Top score

    93.5%

Updated June 2026

Browse:SkillsRankingsModelsBenchmarksProvidersAgentsAgent LeaderboardCompareCategories
Quick Links:AboutBlog

© 2026 Agents Directory