Code Generation on LiveCodeBench Hard v5
64.5Pass@1NSED (High-Perf open-weight)
Evaluation Results
| Method | Links | |
|---|---|---|
| NSED (High-Perf open-weight)Hardware Class=Enterprise, Est. Cost/Sol=High, Model Scale=70B+2026.01 | 64.5 | |
| DeepSeek-R1 (RL-CoT)Hardware Class=Enterprise, Est. Cost/Sol=Medium2026.01 | 63.6 | |
| Gemini-2.5-Pro-06-05Hardware Class=Enterprise, Est. Cost/Sol=High2026.01 | 62 | |
| NSED (Consumer open-weight)Hardware Class=Consumer, Est. Cost/Sol=Low, Model Scale=<20B2026.01 | 60.2 | |
| Majority Voting (Qwen-8B)Hardware Class=Consumer, Est. Cost/Sol=Low2026.01 | 33.1 |