Competitive Programming on LiveCodeBench 2408 - 2505 v6 (Pass@1)
90.7Pass@1Gemini-3 Pro
Evaluation results
| Method | Links | |
|---|---|---|
| Gemini-3 Pro2026.03 | 90.7 | |
| DeepSeek-v3.2-Speciale2026.03 | 88.7 | |
| Nemotron-Cascade-2-30B-A3BThinking budget=128K, Sampling temperature=1.0, top_p=0.95, TIR=true2026.03 | 88.4 | |
| Nemotron-Cascade-2-30B-A3BThinking budget=128K, Sampling temperature=1.0, top_p=0.952026.03 | 87.2 | |
| GPT-OSS-120Bconfig=high2026.03 | 87 | |
| Kimi-K2.5-1T-thinking2026.03 | 85 | |
| Qwen-3.5-397B-A17B2026.03 | 83.6 | |
| GPT-o4-miniconfig=high2026.03 | 80.2 | |
| Qwen-3.5-122B-A10B2026.03 | 78.9 | |
| Nemotron-3-Super-120B-A12B2026.03 | 78.7 | |
| Qwen3-235B-A22B-Thinking-25072026.03 | 78.7 | |
| Qwen-3.5-35B-A3B2026.03 | 74.6 | |
| Nemotron-Cascade-14B2026.03 | 74.6 | |
| Qwen3-Next-80B-A3B-Thinking2026.03 | 73.2 | |
| Nemotron-3-Nano-30B-A3B2026.03 | 68.3 |