Views
No views yet
| Parameter | Value |
|---|---|
| temperature | 0.6 |
| top_p | 0.95 |
| top_k | 20 (or -1 to disable) |
| max_new_tokens | 32768 |
1from transformers import AutoModelForCausalLM, AutoTokenizer
2
3model_name = "IIGroup/X-Coder-RL-Qwen2.5-7B"
4tokenizer = AutoTokenizer.from_pretrained(model_name)
5model = AutoModelForCausalLM.from_pretrained(model_name, device_map="auto")
6
7prompt = "Write a Python function to solve the two sum problem."
8inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
9outputs = model.generate(
10 **inputs,
11 max_new_tokens=32768,
12 temperature=0.6,
13 top_p=0.95,
14 top_k=20,
15 do_sample=True
16)
17print(tokenizer.decode(outputs[0], skip_special_tokens=True))1@misc{wu2026xcoderadvancingcompetitiveprogramming,
2 title={X-Coder: Advancing Competitive Programming with Fully Synthetic Tasks, Solutions, and Tests},
3 author={Jie Wu and Haoling Li and Xin Zhang and Jiani Guo and Jane Luo and Steven Liu and Yangyu Huang and Ruihang Chu and Scarlett Li and Yujiu Yang},
4 year={2026},
5 eprint={2601.06953},
6 archivePrefix={arXiv},
7 primaryClass={cs.CL},
8 url={https://arxiv.org/abs/2601.06953},
9}