Code Generation on LiveCodeBench 2025/01/01 - 2025/05/01 v6 (test)
72.53AccuracyTeam-of-Thoughts
Evaluation Results
| Method | Links | |
|---|---|---|
| Team-of-Thoughtsorchestrator=GPT-5 Mini, context window=16,384 tokens2026.02 | 72.53 | |
| Gemini 3 Flashcontext window=4,096 tokens2026.02 | 70.33 | |
| GPT-5 Minicontext window=4,096 tokens2026.02 | 65.93 | |
| DeepSeek v3.2context window=4,096 tokens2026.02 | 65.93 | |
| Qwen3-vl 235Bcontext window=4,096 tokens2026.02 | 65.93 | |
| AgentVersecontext window=4,096 tokens2026.02 | 65.93 | |
| Claude Sonnet 4.5context window=4,096 tokens2026.02 | 51.1 | |
| GPT-OSS 20Bcontext window=4,096 tokens2026.02 | 50 | |
| Phi-4context window=4,096 tokens2026.02 | 24.18 |