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

---

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

Using llama.cpp commit eca0fab for quantization.

Original model: Qwen/Qwen2.5-3B-Instruct

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	755	112.0612	12.81	8.02	0.97138
IQ1_M	811	42.7456	13.76	21.03	0.34718
IQ2_XXS	905	25.2117	15.36	35.65	0.20222
IQ2_XS	984	15.9149	16.7	56.48	0.11965
IQ2_S	1013	14.5975	17.19	61.58	0.1082
IQ2_M	1088	12.8779	18.46	69.8	0.09436
Q2_K_S	1143	13.0878	19.4	68.68	0.09636
Q2_K	1216	11.8001	20.63	76.18	0.08674
IQ3_XXS	1224	10.6049	20.77	84.76	0.07572
IQ3_XS	1328	10.0306	22.54	89.61	0.06975
Q3_K_S	1387	15.5457	23.54	57.82	0.11941
IQ3_S	1390	9.9591	23.59	90.26	0.06984
IQ3_M	1420	9.9957	24.1	89.93	0.06962
Q3_K_M	1517	14.0989	25.74	63.76	0.10568
Q3_K_L	1629	13.8579	27.64	64.86	0.10372
IQ4_XS	1659	9.2935	28.15	96.72	0.06517
IQ4_NL	1741	9.2824	29.54	96.84	0.06503
Q4_0	1744	9.485	29.59	94.77	0.06626
Q4_K_S	1750	9.2573	29.7	97.1	0.06485
Q4_K_M	1841	9.2305	31.24	97.38	0.06475
Q4_1	1904	9.2746	32.31	96.92	0.06512
Q5_K_S	2070	9.1338	35.13	98.41	0.06402
Q5_0	2075	9.1513	35.21	98.22	0.06413
Q5_K_M	2122	9.1339	36.01	98.41	0.06407
Q5_1	2235	9.1231	37.93	98.53	0.06386
Q6_K	2421	9.069	41.08	99.12	0.06342
Q8_0	3134	9.0114	53.18	99.75	0.06285
F16	5893	8.9888	100	100	0.06268

---

Qwen2.5-3B-Instruct

Introduction

Qwen2.5 is the latest series of Qwen large language models. For Qwen2.5, we release a number of base language models and instruction-tuned language models ranging from 0.5 to 72 billion parameters. Qwen2.5 brings the following improvements upon Qwen2:

• Significantly more knowledge and has greatly improved capabilities in coding and mathematics, thanks to our specialized expert models in these domains.
• Significant improvements in instruction following, generating long texts (over 8K tokens), understanding structured data (e.g, tables), and generating structured outputs especially JSON. More resilient to the diversity of system prompts, enhancing role-play implementation and condition-setting for chatbots.
• Long-context Support up to 128K tokens and can generate up to 8K tokens.
• Multilingual support for over 29 languages, including Chinese, English, French, Spanish, Portuguese, German, Italian, Russian, Japanese, Korean, Vietnamese, Thai, Arabic, and more.

This repo contains the instruction-tuned 3B Qwen2.5 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 and generation 8192 tokens

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

Requirements

The code of Qwen2.5 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/Qwen2.5-3B-Instruct"

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

prompt = "Give me a short introduction to large language model."
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.

@misc{qwen2.5,
    title = {Qwen2.5: A Party of Foundation Models},
    url = {https://qwenlm.github.io/blog/qwen2.5/},
    author = {Qwen Team},
    month = {September},
    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}
}