ThomasBaruzier/Qwen2.5-Coder-3B-Instruct-GGUF

Llama.cpp imatrix quantizations of Qwen/Qwen2.5-Coder-3B-Instruct

Using llama.cpp commit 3ad5451 for quantization.

All quants were made using the imatrix option and Bartowski's calibration file.

---

Perplexity table (the lower the better)

Quant	Size (MB)	PPL	Size (%)	Accuracy (%)	PPL error rate
IQ1_S	754	38.2733	12.80	25.21	0.68
IQ1_M	810	23.9542	13.75	40.28	0.41
IQ2_XXS	904	22.7872	15.34	42.34	0.41
IQ2_XS	983	14.4025	16.68	66.99	0.24
IQ2_S	1012	13.6118	17.17	70.88	0.22
IQ2_M	1087	12.6560	18.45	76.24	0.20
Q2_K_S	1142	12.4745	19.38	77.35	0.20
Q2_K	1215	11.5168	20.62	83.78	0.18
IQ3_XXS	1223	10.8327	20.76	89.07	0.17
IQ3_XS	1327	10.2693	22.52	93.96	0.16
Q3_K_S	1386	32.8121	23.52	29.41	0.62
IQ3_S	1389	10.2409	23.57	94.22	0.16
IQ3_M	1419	10.1950	24.08	94.64	0.16
Q3_K_M	1516	20.7076	25.73	46.59	0.36
Q3_K_L	1628	20.6384	27.63	46.75	0.36
IQ4_XS	1658	9.9013	28.14	97.45	0.15
IQ4_NL	1740	9.8926	29.53	97.53	0.15
Q4_0	1743	10.0757	29.58	95.76	0.16
Q4_K_S	1749	9.8497	29.68	97.96	0.15
Q4_K_M	1840	9.8160	31.23	98.29	0.15
Q4_1	1903	9.8176	32.30	98.28	0.15
Q5_K_S	2069	9.7456	35.12	99.00	0.15
Q5_0	2074	9.7237	35.20	99.23	0.15
Q5_K_M	2121	9.7331	36.00	99.13	0.15
Q5_1	2234	9.7260	37.92	99.20	0.15
Q6_K	2420	9.6939	41.07	99.53	0.15
Q8_0	3133	9.6710	53.17	99.77	0.15
F16	5892	9.6486	100	100	0.15

---

Qwen2.5-Coder-3B-Instruct

Introduction

Qwen2.5-Coder is the latest series of Code-Specific Qwen large language models (formerly known as CodeQwen). As of now, Qwen2.5-Coder has covered six mainstream model sizes, 0.5, 1.5, 3, 7, 14, 32 billion parameters, to meet the needs of different developers. Qwen2.5-Coder brings the following improvements upon CodeQwen1.5:

• Significantly improvements in code generation, code reasoning and code fixing. Base on the strong Qwen2.5, we scale up the training tokens into 5.5 trillion including source code, text-code grounding, Synthetic data, etc. Qwen2.5-Coder-32B has become the current state-of-the-art open-source codeLLM, with its coding abilities matching those of GPT-4o.
• A more comprehensive foundation for real-world applications such as Code Agents. Not only enhancing coding capabilities but also maintaining its strengths in mathematics and general competencies.

This repo contains the instruction-tuned 3B Qwen2.5-Coder model, which has the following features:

• Type: Causal Language Models
• Training Stage: Pretraining & Post-training
• Architecture: transformers with RoPE, SwiGLU, RMSNorm, Attention QKV bias and tied word embeddings
• Number of Parameters: 3.09B
• Number of Paramaters (Non-Embedding): 2.77B
• Number of Layers: 36
• Number of Attention Heads (GQA): 16 for Q and 2 for KV
• Context Length: Full 32,768 tokens

For more details, please refer to our blog, GitHub, Documentation, Arxiv.

Requirements

The code of Qwen2.5-Coder has been in the latest Hugging face transformers and we advise you to use the latest version of transformers.

With transformers<4.37.0, you will encounter the following error:

KeyError: 'qwen2'

Quickstart

Here provides a code snippet with apply_chat_template to show you how to load the tokenizer and model and how to generate contents.

from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "Qwen/Qwen2.5-Coder-3B-Instruct"

model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype="auto",
    device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained(model_name)

prompt = "write a quick sort algorithm."
messages = [
    {"role": "system", "content": "You are Qwen, created by Alibaba Cloud. You are a helpful assistant."},
    {"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)

generated_ids = model.generate(
    **model_inputs,
    max_new_tokens=512
)
generated_ids = [
    output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]

response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]

Evaluation & Performance

Detailed evaluation results are reported in this 📑 blog.

For requirements on GPU memory and the respective throughput, see results here.

Citation

If you find our work helpful, feel free to give us a cite.

@article{hui2024qwen2,
      title={Qwen2. 5-Coder Technical Report},
      author={Hui, Binyuan and Yang, Jian and Cui, Zeyu and Yang, Jiaxi and Liu, Dayiheng and Zhang, Lei and Liu, Tianyu and Zhang, Jiajun and Yu, Bowen and Dang, Kai and others},
      journal={arXiv preprint arXiv:2409.12186},
      year={2024}
}
@article{qwen2,
      title={Qwen2 Technical Report}, 
      author={An Yang and Baosong Yang and Binyuan Hui and Bo Zheng and Bowen Yu and Chang Zhou and Chengpeng Li and Chengyuan Li and Dayiheng Liu and Fei Huang and Guanting Dong and Haoran Wei and Huan Lin and Jialong Tang and Jialin Wang and Jian Yang and Jianhong Tu and Jianwei Zhang and Jianxin Ma and Jin Xu and Jingren Zhou and Jinze Bai and Jinzheng He and Junyang Lin and Kai Dang and Keming Lu and Keqin Chen and Kexin Yang and Mei Li and Mingfeng Xue and Na Ni and Pei Zhang and Peng Wang and Ru Peng and Rui Men and Ruize Gao and Runji Lin and Shijie Wang and Shuai Bai and Sinan Tan and Tianhang Zhu and Tianhao Li and Tianyu Liu and Wenbin Ge and Xiaodong Deng and Xiaohuan Zhou and Xingzhang Ren and Xinyu Zhang and Xipin Wei and Xuancheng Ren and Yang Fan and Yang Yao and Yichang Zhang and Yu Wan and Yunfei Chu and Yuqiong Liu and Zeyu Cui and Zhenru Zhang and Zhihao Fan},
      journal={arXiv preprint arXiv:2407.10671},
      year={2024}
}