ThomasBaruzier/Qwen2.5-1.5B-Instruct-GGUF

---

Llama.cpp imatrix quantizations of Qwen/Qwen2.5-1.5B-Instruct

Using llama.cpp commit eca0fab for quantization.

Original model: Qwen/Qwen2.5-1.5B-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	417	193.6245	14.13	5.24	1.77149
IQ1_M	443	66.9068	15.01	15.17	0.52878
IQ2_XXS	488	33.3356	16.54	30.45	0.25559
IQ2_XS	525	20.287	17.79	50.04	0.14936
IQ2_S	538	18.2927	18.23	55.49	0.1338
IQ2_M	574	15.4838	19.45	65.56	0.11113
Q2_K_S	611	16.0169	20.7	63.38	0.11623
IQ3_XXS	638	12.3935	21.62	81.91	0.0877
Q2_K	645	14.1657	21.86	71.66	0.10105
IQ3_XS	698	11.7112	23.65	86.68	0.08256
Q3_K_S	726	12.4782	24.6	81.35	0.08842
IQ3_S	728	11.4241	24.67	88.86	0.07977
IQ3_M	741	11.4058	25.11	89	0.07862
Q3_K_M	786	11.3529	26.64	89.42	0.08018
Q3_K_L	840	11.1934	28.46	90.69	0.07913
IQ4_XS	855	10.5302	28.97	96.4	0.07351
IQ4_NL	893	10.5116	30.26	96.57	0.07335
Q4_0	895	10.8217	30.33	93.8	0.07576
Q4_K_S	897	10.5236	30.4	96.46	0.0736
Q4_K_M	941	10.4628	31.89	97.02	0.0731
Q4_1	970	10.51	32.87	96.59	0.07347
Q5_K_S	1048	10.2715	35.51	98.83	0.07148
Q5_0	1051	10.3196	35.62	98.37	0.07212
Q5_K_M	1073	10.2529	36.36	99.01	0.07143
Q5_1	1126	10.2624	38.16	98.92	0.0714
Q6_K	1214	10.203	41.14	99.49	0.07108
Q8_0	1571	10.167	53.24	99.84	0.07068
F16	2951	10.1512	100	100	0.07058

---

Qwen2.5-1.5B-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 1.5B 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: 1.54B
• Number of Paramaters (Non-Embedding): 1.31B
• Number of Layers: 28
• Number of Attention Heads (GQA): 12 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/Qwen2.5-1.5B-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}
}