ThomasBaruzier/Qwen2.5-32B-Instruct-GGUF

---

Llama.cpp imatrix quantizations of Qwen/Qwen2.5-32B-Instruct

Using llama.cpp commit eca0fab for quantization.

Original model: Qwen/Qwen2.5-32B-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	6938	12.2991	11.1	44.87	0.08384
IQ1_M	7565	10.2638	12.1	53.77	0.0699
IQ2_XS	9497	7.3601	15.2	74.99	0.04846
IQ2_S	9907	7.2397	15.85	76.23	0.04762
IQ2_M	10743	6.7268	17.19	82.05	0.04354
Q2_K_S	10956	6.9981	17.53	78.87	0.04644
Q2_K	11743	6.6603	18.79	82.87	0.04324
IQ3_XXS	12245	6.157	19.59	89.64	0.03929
IQ3_XS	13071	6.0366	20.91	91.43	0.03833
Q3_K_S	13726	6.0878	21.96	90.66	0.03872
IQ3_S	13769	5.9886	22.03	92.16	0.03816
IQ3_M	14125	5.9942	22.6	92.07	0.03802
Q3_K_M	15197	5.8008	24.32	95.14	0.03677
Q3_K_L	16449	5.7812	26.32	95.47	0.03667
IQ4_XS	16874	5.6502	27	97.68	0.03586
IQ4_NL	17817	5.6408	28.51	97.84	0.03575
Q4_0	17845	5.6946	28.55	96.92	0.03599
Q4_K_S	17915	5.6367	28.66	97.91	0.03561
Q4_K_M	18932	5.6224	30.29	98.16	0.03554
IQ2_XXS	8611	8.0187	13.78	68.83	0.05388
Q4_1	19684	5.6586	31.49	97.53	0.03587
Q5_K_S	21590	5.568	34.54	99.12	0.03515
Q5_0	21658	5.588	34.65	98.77	0.03538
Q5_K_M	22185	5.567	35.5	99.14	0.03515
Q5_1	23496	5.5734	37.59	99.03	0.0352
Q6_K	25641	5.5305	41.03	99.79	0.03483
Q8_0	33208	5.5221	53.13	99.95	0.03478
F16	62500	5.5191	100	100	0.03474

---

Qwen2.5-32B-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 32B Qwen2.5 model, which has the following features:

• Type: Causal Language Models
• Training Stage: Pretraining & Post-training
• Architecture: transformers with RoPE, SwiGLU, RMSNorm, and Attention QKV bias
• Number of Parameters: 32.5B
• Number of Paramaters (Non-Embedding): 31.0B
• Number of Layers: 64
• Number of Attention Heads (GQA): 40 for Q and 8 for KV
• Context Length: Full 131,072 tokens and generation 8192 tokens
  • Please refer to this section for detailed instructions on how to deploy Qwen2.5 for handling long texts.

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/Qwen2.5-32B-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]

Processing Long Texts

The current config.json is set for context length up to 32,768 tokens. To handle extensive inputs exceeding 32,768 tokens, we utilize YaRN, a technique for enhancing model length extrapolation, ensuring optimal performance on lengthy texts.

For supported frameworks, you could add the following to config.json to enable YaRN:

{
  ...,
  "rope_scaling": {
    "factor": 4.0,
    "original_max_position_embeddings": 32768,
    "type": "yarn"
  }
}

For deployment, we recommend using vLLM. Please refer to our Documentation for usage if you are not familar with vLLM. Presently, vLLM only supports static YARN, which means the scaling factor remains constant regardless of input length, potentially impacting performance on shorter texts. We advise adding the rope_scaling configuration only when processing long contexts is required.

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}
}