Codegolf v2.2
text
+
+
+
+
About
Codegolf v2.2 is a specialized programming benchmark that evaluates AI models' ability to write extremely concise code while maintaining correctness and functionality. This benchmark tests code optimization skills, creative problem-solving, and deep programming language understanding through challenges that require minimal character count solutions. Codegolf v2.2 measures AI systems' capacity for elegant, efficient code generation under strict brevity constraints.
+
+
+
+
Evaluation Stats
Total Models4
Organizations1
Verified Results0
Self-Reported4
+
+
+
+
Benchmark Details
Max Score1
Language
en
+
+
+
+
Performance Overview
Score distribution and top performers
Score Distribution
4 models
Top Score
16.8%
Average Score
13.9%
High Performers (80%+)
0Top Organizations
#1Google
4 models
13.9%
+
+
+
+
Leaderboard
4 models ranked by performance on Codegolf v2.2
License | Links | ||||
---|---|---|---|---|---|
Jun 26, 2025 | Proprietary | 16.8% | |||
May 20, 2025 | Gemma | 16.8% | |||
Jun 26, 2025 | Proprietary | 11.0% | |||
May 20, 2025 | Gemma | 11.0% |