Instructions to use Qwen/Qwen2-72B-Instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Qwen/Qwen2-72B-Instruct with Transformers:

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

pipe = pipeline("text-generation", model="Qwen/Qwen2-72B-Instruct")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen2-72B-Instruct")
model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2-72B-Instruct")
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

vLLM

How to use Qwen/Qwen2-72B-Instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Qwen/Qwen2-72B-Instruct"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Qwen/Qwen2-72B-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Qwen/Qwen2-72B-Instruct

SGLang

How to use Qwen/Qwen2-72B-Instruct 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 "Qwen/Qwen2-72B-Instruct" \
    --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": "Qwen/Qwen2-72B-Instruct",
		"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 "Qwen/Qwen2-72B-Instruct" \
        --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": "Qwen/Qwen2-72B-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Qwen/Qwen2-72B-Instruct with Docker Model Runner:
```
docker model run hf.co/Qwen/Qwen2-72B-Instruct
```

使用128k长度运行报错，只要6万字长度

#17

by loong - opened Jun 29, 2024

Discussion

loong

Jun 29, 2024

BadRequestError: Error code: 400 - {'object': 'error', 'message': "This model's maximum context length is 32768 tokens. However, you requested 37055 tokens in the messages, Please reduce the length of the messages.", 'type': 'BadRequestError', 'param': None, 'code': 400}

ming1114

Jun 29, 2024

运行命令发一下

loong

Jul 1, 2024

用的vllm推理，修改了config：
python -m vllm.entrypoints.openai.api_server --model /ai/qwen2-7b-yarn --host 192.1**** --port 10860 --max-model-len 32768 --trust-remote-code --tensor-parallel-size 1 --dtype=half

loong

Jul 1, 2024

•

edited Jul 1, 2024

大概测试下来4万多字可以，再往上就报前面错误了 @ming1114

loong

Jul 1, 2024

我也在github上提了，https://github.com/QwenLM/Qwen2/issues/717

donglai1

Jul 3, 2024

因为你运行命令中是设置为： --max-model-len 32768，你超过这个长度肯定报错，你不设置，默认应该是128k ，但是推理服务不一定能加载启动得了，看你显存大小。在显存还算够用能加载启动得情况下，增加这个值，看看

aabbccddwasd

Jul 4, 2024

增加这个就告诉我最大embeddings是32768怎么办啊

soikit

Nov 6, 2024

•

edited Nov 6, 2024

我是想让一次对话处理超过33000（token）的文本，然后我一开始在没有修改max_position_embeddings时，也报错了，在config.json中加了rope_scaling相关的配置也报错，后来我只是将max_position_embeddings从32768增加至64768，然后再处理这个33000(token)长的文本时，就可以正常处理了。

但是我也不知道修改这个max_position_embeddings会有什么样的后果（如果有知道的大佬，欢迎告知啊）

哦，我是用swift+lmdeploy部署的32B模型，部署的命令是：CUDA_VISIBLE_DEVICES=0,1,2,3 swift deploy --model_type qwen2_5-32b-instruct --model_id_or_path /root/autodl-fs/Qwen/Qwen2___5-32B-Instruct --infer_backend lmdeploy --tp 4 --port 6006 --max_length 512000 --dtype bf16

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment