metadata

license: apache-2.0
language:
  - th
  - zh
  - en
metrics:
  - accuracy
base_model:
  - Qwen/Qwen2.5-7B
pipeline_tag: text-generation
tags:
  - chemistry
  - biology
  - finance
  - legal
  - code
  - medical
  - text-generation-inference

OpenThaiLLM-: Thai & China Large Language Model (Instruct)

OpenThaiLLM-DoodNiLT-Instruct is an 7 billion parameter instruct model designed for Thai 🇹🇭 & China 🇨🇳 language. It demonstrates competitive performance with GPT-3.5-turbo and llama-3-typhoon-v1.5-8b-instruct, and is optimized for application use cases, Retrieval-Augmented Generation (RAG), constrained generation, and reasoning tasks.is a Thai 🇹🇭 & China 🇨🇳 large language model with 7 billion parameters, and it is based on Qwen2-7B.

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 base 7B Qwen2.5 model, which has the following features:

Type: Causal Language Models
Training Stage: Pretraining
Architecture: transformers with RoPE, SwiGLU, RMSNorm, and Attention QKV bias
Number of Parameters: 7.61B
Number of Paramaters (Non-Embedding): 6.53B
Number of Layers: 28
Number of Attention Heads (GQA): 28 for Q and 4 for KV
Context Length: 131,072 tokens

We do not recommend using base language models for conversations. Instead, you can apply post-training, e.g., SFT, RLHF, continued pretraining, etc., on this model.

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'

## Training details

We pretrained the models with a large amount of data, and we post-trained the models with both supervised finetuning and direct preference optimization.

## Implementation

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

```python
from transformers import AutoModelForCausalLM, AutoTokenizer
device = "cuda" # the device to load the model onto

model = AutoModelForCausalLM.from_pretrained(
    "nectec/OpenThaiLLM-DoodNiLT-V1.0.0-Beta-7B-Instruct",
    torch_dtype="auto",
    device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained("nectec/OpenThaiLLM-DoodNiLT-V1.0.0-Beta-7B-Instruct")

prompt = "บริษัท A มีต้นทุนคงที่ 100,000 บาท และต้นทุนผันแปรต่อหน่วย 50 บาท ขายสินค้าได้ในราคา 150 บาทต่อหน่วย ต้องขายสินค้าอย่างน้อยกี่หน่วยเพื่อให้ถึงจุดคุ้มทุน?"
messages = [
    {"role": "system", "content": "คุณคือ DoodNiLT Assistant จงตอบคำถามอธิบายเป็นภาษาไทย"},
    {"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.input_ids,
    max_new_tokens=4096,
    repetition_penalty=1.2
)
response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
print(response)

Evaluation Performance

Model	ONET	IC	TGAT	TPAT-1	A-Level	Average (ThaiExam)	M3Exam (1 shot)	MMLU
OpenthaiLLM-Prebuilt-7B	0.5185	0.6421	0.6461	0.4224	0.3937	0.5245	0.5355	0.6644
SeaLLM-v3-7B	0.4753	0.6421	0.6153	0.3275	0.3464	0.4813	0.7037	0.4907
llama-3-typhoon-v1.5-8B	0.3765	0.3473	0.5538	0.4137	0.2913	0.3965	0.6451	0.4312
Qwen-2-7B	0.4814	0.621	0.6153	0.3448	0.3385	0.4802	0.7073	0.4949
Meta-Llama-3.1-8B	0.3641	0.2631	0.2769	0.3793	0.1811	0.2929	0.6591	0.4239

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}
}