Code Generation on AetherCode (AC)
54.25pass@1Claude Code
Evaluation Results
| Method | Links | |
|---|---|---|
| Claude CodeBackbone=Claude Opus 4.62026.05 | 54.25 | |
| SolvitaBackbone=Claude Opus 4.62026.05 | 53.75 | |
| SolvitaBackbone=DeepSeek V4 Pro2026.05 | 51.5 | |
| SolvitaBackbone=GPT-5.42026.05 | 49.25 | |
| Codex CLIBackbone=GPT-5.42026.05 | 48.5 | |
| Codex CLIBackbone=DeepSeek V4 Pro2026.05 | 45 | |
| Codex CLIBackbone=Claude Opus 4.62026.05 | 44.25 | |
| Claude CodeBackbone=DeepSeek V4 Pro2026.05 | 43.75 | |
| Claude CodeBackbone=GPT-5.42026.05 | 42.75 | |
| MapCoderBackbone=Claude Opus 4.62026.05 | 38.25 | |
| AlphaCodiumBackbone=Claude Opus 4.62026.05 | 36.5 | |
| MapCoderBackbone=DeepSeek V4 Pro2026.05 | 35 | |
| SolvitaBackbone=Grok2026.05 | 33.5 | |
| AlphaCodiumBackbone=GPT-5.42026.05 | 33 | |
| AlphaCodiumBackbone=DeepSeek V4 Pro2026.05 | 33 | |
| Codex CLIBackbone=Grok2026.05 | 31.5 | |
| MapCoderBackbone=GPT-5.42026.05 | 30.5 | |
| Claude CodeBackbone=Grok2026.05 | 29.75 | |
| SolvitaBackbone=Qwen3.62026.05 | 26 | |
| Single-passBackbone=DeepSeek V4 Pro2026.05 | 24 | |
| Single-passBackbone=Claude Opus 4.62026.05 | 22.75 | |
| Codex CLIBackbone=Qwen3.62026.05 | 22.5 | |
| Claude CodeBackbone=Qwen3.62026.05 | 21.5 | |
| AlphaCodiumBackbone=Grok2026.05 | 21 | |
| MapCoderBackbone=Grok2026.05 | 19.5 | |
| Single-passBackbone=GPT-5.42026.05 | 18 | |
| MapCoderBackbone=Qwen3.62026.05 | 16 | |
| Single-passBackbone=Grok2026.05 | 15.5 | |
| AlphaCodiumBackbone=Qwen3.62026.05 | 14.5 | |
| Single-passBackbone=Qwen3.62026.05 | 9.5 |