LiveCodeBench v6
text
+
+
+
+
About
LiveCodeBench v6 is the latest iteration of the LiveCodeBench framework featuring advanced evaluation methodologies and expanded problem diversity from competitive programming platforms. This version includes enhanced metrics for code quality, efficiency assessment, and comprehensive testing of AI models' programming capabilities through continuously updated, contamination-free coding challenges.
+
+
+
+
Evaluation Stats
Total Models7
Organizations3
Verified Results0
Self-Reported7
+
+
+
+
Benchmark Details
Max Score1
Language
en
+
+
+
+
Performance Overview
Score distribution and top performers
Score Distribution
7 models
Top Score
82.8%
Average Score
59.1%
High Performers (80%+)
1Top Organizations
#1Zhipu AI
1 model
82.8%
#2Alibaba Cloud / Qwen Team
4 models
62.8%
#3Moonshot AI
2 models
40.0%
+
+
+
+
Leaderboard
7 models ranked by performance on LiveCodeBench v6
| License | Links | ||||
|---|---|---|---|---|---|
| Sep 30, 2025 | MIT | 82.8% | |||
| Jul 25, 2025 | Apache 2.0 | 74.1% | |||
| Sep 10, 2025 | Apache 2.0 | 68.7% | |||
| Sep 10, 2025 | Apache 2.0 | 56.6% | |||
| Jul 11, 2025 | MIT | 53.7% | |||
| Jul 22, 2025 | Apache 2.0 | 51.8% | |||
| Jul 11, 2025 | MIT | 26.3% |
