Queueing-simulation code generation on Queueing-simulation held-out (test)
86Executability RateQwen2.5-Coder-7B
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Qwen2.5-Coder-7BTraining Status=Fine-tuned2026.01 | 86 | 86 | 76.8 | |
| Qwen2.5-Coder-7BTraining Status=Pre-trained2026.01 | 80.4 | 80.4 | 0 | |
| DeepSeek-Coder-6.7BTraining Status=Fine-tuned2026.01 | 75 | 74.8 | 62.3 | |
| DeepSeek-Coder-6.7BTraining Status=Pre-trained2026.01 | 26.1 | 0.09 | 0 |