Code Generation on HumanEval (TSR, P-ASR, T-ASR, R-ASR)
90TSRClaude Code
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Claude CodeInjection Strategy=None, Backend Model=Qwen 3.6 Plus2026.05 | 90 | — | — | — | |
| Claude Code (via EDIT S.)Injection Strategy=via EDIT S., Backend Model=Qwen 3.6 Plus2026.05 | 90 | 100 | 100 | 25 | |
| OpenCodeInjection Strategy=None, Backend Model=Qwen 3.6 Plus2026.05 | 85 | — | — | — | |
| Claude Code (via ADD S.)Injection Strategy=via ADD S., Backend Model=Qwen 3.6 Plus2026.05 | 80 | 55 | 55 | 55 | |
| Kilo CodeInjection Strategy=None, Backend Model=Qwen 3.6 Plus2026.05 | 80 | — | — | — | |
| Kilo Code (via ADD S.)Injection Strategy=via ADD S., Backend Model=Qwen 3.6 Plus2026.05 | 80 | 100 | 100 | 100 | |
| OpenCode (via EDIT S.)Injection Strategy=via EDIT S., Backend Model=Qwen 3.6 Plus2026.05 | 75 | 90 | 90 | 45 | |
| Kilo Code (via EDIT S.)Injection Strategy=via EDIT S., Backend Model=Qwen 3.6 Plus2026.05 | 75 | 95 | 95 | 75 | |
| OpenCode (via ADD S.)Injection Strategy=via ADD S., Backend Model=Qwen 3.6 Plus2026.05 | 70 | 100 | 100 | 100 |