ThomasBaruzier/Qwen2.5-0.5B-Instruct-GGUF

---

Llama.cpp imatrix quantizations of Qwen/Qwen2.5-0.5B-Instruct

Using llama.cpp commit eca0fab for quantization.

Original model: Qwen/Qwen2.5-0.5B-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	302	30.0453	31.82	50.53	0.23714
IQ1_M	304	24.9151	32.03	60.93	0.19238
IQ2_XXS	307	21.4864	32.35	70.66	0.16704
IQ2_XS	310	19.7829	32.67	76.74	0.15355
IQ2_S	311	19.2041	32.77	79.06	0.14841
IQ2_M	314	18.325	33.09	82.85	0.14001
Q2_K_S	316	18.3462	33.3	82.75	0.14091
IQ3_XXS	319	16.9784	33.61	89.42	0.12828
Q3_K_S	323	17.7765	34.04	85.4	0.13668
IQ3_S	323	16.3794	34.04	92.69	0.12173
IQ3_XS	323	16.3794	34.04	92.69	0.12173
Q2_K	323	17.6841	34.04	85.85	0.13561
IQ3_M	327	16.3667	34.46	92.76	0.12182
IQ4_XS	334	15.8792	35.19	95.61	0.11933
IQ4_NL	337	15.8468	35.51	95.8	0.11921
Q4_0	337	17.1007	35.51	88.78	0.13053
Q3_K_M	339	15.8499	35.72	95.79	0.11934
Q3_K_L	353	15.7298	37.2	96.52	0.1182
Q4_1	358	16.1819	37.72	93.82	0.12328
Q4_K_S	368	15.5497	38.78	97.63	0.11716
Q5_0	380	15.5038	40.04	97.92	0.11702
Q4_K_M	380	15.4428	40.04	98.31	0.11637
Q5_K_S	394	15.5266	41.52	97.78	0.11682
Q5_1	400	15.4875	42.15	98.03	0.11641
Q5_K_M	401	15.4788	42.26	98.08	0.11631
Q6_K	483	15.2145	50.9	99.79	0.11422
Q8_0	507	15.239	53.42	99.63	0.11452
F16	949	15.1819	100	100	0.114

---

Qwen2.5-0.5B-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 0.5B 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: 0.49B
• Number of Paramaters (Non-Embedding): 0.36B
• Number of Layers: 24
• Number of Attention Heads (GQA): 14 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/Qwen2.5-0.5B-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}
}