Instructions to use rinna/qwen2.5-bakeneko-32b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use rinna/qwen2.5-bakeneko-32b with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="rinna/qwen2.5-bakeneko-32b")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("rinna/qwen2.5-bakeneko-32b")
model = AutoModelForCausalLM.from_pretrained("rinna/qwen2.5-bakeneko-32b", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use rinna/qwen2.5-bakeneko-32b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "rinna/qwen2.5-bakeneko-32b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "rinna/qwen2.5-bakeneko-32b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/rinna/qwen2.5-bakeneko-32b

SGLang

How to use rinna/qwen2.5-bakeneko-32b with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "rinna/qwen2.5-bakeneko-32b" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "rinna/qwen2.5-bakeneko-32b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "rinna/qwen2.5-bakeneko-32b" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "rinna/qwen2.5-bakeneko-32b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use rinna/qwen2.5-bakeneko-32b with Docker Model Runner:
```
docker model run hf.co/rinna/qwen2.5-bakeneko-32b
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

`Qwen2.5 Bakeneko 32B (rinna/qwen2.5-bakeneko-32b)`

Overview

We conduct continual pre-training of Qwen/Qwen2.5-32B on 18B tokens from a mixture of Japanese and English datasets. The continual pre-training improves the model's performance on Japanese tasks.

The name bakeneko comes from the Japanese word 化け猫/ばけねこ/Bakeneko, which is a kind of Japanese mythical creature (妖怪/ようかい/Youkai).

Model Type	Model Name
Japanese Continual Pre-Training Model	Qwen2.5 Bakeneko 32B [HF]
Instruction-Tuning Model	Qwen2.5 Bakeneko 32B Instruct [HF][AWQ][GGUF][GPTQ int8][GPTQ int4]
DeepSeek R1 Distill Qwen2.5 Merged Reasoning Model	DeepSeek R1 Distill Qwen2.5 Bakeneko 32B [HF][AWQ][GGUF][GPTQ int8][GPTQ int4]
QwQ Merged Reasoning Model	QwQ Bakeneko 32B [HF][AWQ][GGUF][GPTQ int8][GPTQ int4]
QwQ Bakeneko Merged Instruction-Tuning Model	Qwen2.5 Bakeneko 32B Instruct V2 [HF][AWQ][GGUF][GPTQ int8][GPTQ int4]

Library

The model was trained using code based on Lightning-AI/litgpt.
Model architecture

A 64-layer, 5120-hidden-size transformer-based language model. Please refer to the Qwen2.5 Technical Report for detailed information on the model's architecture.
Training

The model was initialized with the Qwen/Qwen2.5-32B model and continually trained on around 18B tokens from a mixture of the following corpora
- Japanese CC-100
- Japanese C4
- Japanese OSCAR
- The Pile
- Wikipedia
- rinna curated Japanese dataset
Contributors
Release date

February 13, 2025

Benchmarking

Model	Japanese LM Evaluation Harness	Japanese MT-Bench (first turn)	Japanese MT-Bench (multi turn)
Qwen/Qwen2.5-32B	79.46	-	-
rinna/qwen2.5-bakeneko-32b	79.18	-	-
Qwen/Qwen2.5-32B-Instruct	78.29	8.13	7.54
rinna/qwen2.5-bakeneko-32b-instruct	79.62	8.17	7.66
rinna/qwen2.5-bakeneko-32b-instruct-v2	77.92	8.86	8.53
deepseek-ai/DeepSeek-R1-Distill-Qwen-32B	73.51	7.39	6.88
rinna/deepseek-r1-distill-qwen2.5-bakeneko-32b	77.43	8.58	8.19
Qwen/QwQ-32B	76.12	8.58	8.25
rinna/qwq-bakeneko-32b	78.31	8.81	8.52

For detailed benchmarking results, please refer to rinna's LM benchmark page (Sheet 20250213).

Tokenization

The model uses the original Qwen/Qwen2.5-32B tokenizer.

How to cite

@misc{rinna-qwen2.5-bakeneko-32b,
    title = {rinna/qwen2.5-bakeneko-32b},
    author = {Wakatsuki, Toshiaki and Chen, Xinqi and Sawada, Kei},
    url = {https://huggingface.co/rinna/qwen2.5-bakeneko-32b}
}

@inproceedings{sawada2024release,
    title = {Release of Pre-Trained Models for the {J}apanese Language},
    author = {Sawada, Kei and Zhao, Tianyu and Shing, Makoto and Mitsui, Kentaro and Kaga, Akio and Hono, Yukiya and Wakatsuki, Toshiaki and Mitsuda, Koh},
    booktitle = {Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)},
    month = {5},
    year = {2024},
    pages = {13898--13905},
    url = {https://aclanthology.org/2024.lrec-main.1213},
    note = {\url{https://arxiv.org/abs/2404.01657}}
}

References

@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}
}

@misc{litgpt-2023,
    author = {Lightning AI},
    title = {LitGPT},
    howpublished = {\url{https://github.com/Lightning-AI/litgpt}},
    year = {2023}
}