Code Generation on AetherCode LiveCodeBench pro v6 (Pass@1)
2,439Pass@1Gemini 3-pro
Evaluation Results
| Method | Links | |
|---|---|---|
| Gemini 3-proDataset=LiveCodeBench pro2026.03 | 2,439 | |
| GPT-5 HighDataset=LiveCodeBench pro2026.03 | 2,114 | |
| Seed1.8Dataset=LiveCodeBench pro2026.03 | 1,930 | |
| Gemini 2.5-proDataset=LiveCodeBench pro2026.03 | 1,765 | |
| Claude Sonnet-4.5Dataset=LiveCodeBench pro2026.03 | 1,364 | |
| Gemini 3-proDataset=LiveCodeBench(v6)2026.03 | 90.7 | |
| GPT-5 HighDataset=LiveCodeBench(v6)2026.03 | 87 | |
| Seed1.8Dataset=LiveCodeBench(v6)2026.03 | 79.5 | |
| Gemini 2.5-proDataset=LiveCodeBench(v6)2026.03 | 73.6 | |
| Claude Sonnet-4.5Dataset=LiveCodeBench(v6)2026.03 | 64 | |
| Gemini 3-proDataset=AetherCode2026.03 | 56.7 | |
| GPT-5 HighDataset=AetherCode2026.03 | 43.3 | |
| Seed1.8Dataset=AetherCode2026.03 | 38.2 | |
| Gemini 2.5-proDataset=AetherCode2026.03 | 32.4 | |
| Claude Sonnet-4.5Dataset=AetherCode2026.03 | 18.2 |