Instructions to use Qwen/Qwen2.5-Coder-32B-Instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Qwen/Qwen2.5-Coder-32B-Instruct with Transformers:

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

pipe = pipeline("text-generation", model="Qwen/Qwen2.5-Coder-32B-Instruct")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen2.5-Coder-32B-Instruct")
model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2.5-Coder-32B-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
HuggingChat
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use Qwen/Qwen2.5-Coder-32B-Instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Qwen/Qwen2.5-Coder-32B-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/Qwen2.5-Coder-32B-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Qwen/Qwen2.5-Coder-32B-Instruct

SGLang

How to use Qwen/Qwen2.5-Coder-32B-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/Qwen2.5-Coder-32B-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/Qwen2.5-Coder-32B-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/Qwen2.5-Coder-32B-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/Qwen2.5-Coder-32B-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Qwen/Qwen2.5-Coder-32B-Instruct with Docker Model Runner:
```
docker model run hf.co/Qwen/Qwen2.5-Coder-32B-Instruct
```

How to run speculative decoding of this model with 0.5B model.

#18

by Eryk-Chmielewski - opened Nov 20, 2024

Discussion

Eryk-Chmielewski

Nov 20, 2024

How to run speculative decoding of this model with 0.5B model.

I get errors in vllm.

First is that vocab_size is different. So in 0.5B in config.json I set vocab_size=152064 to match with 32B model value.

But now I get error
assert loaded_weight.shape[output_dim] == self.org_vocab_size

I don't see possibility to change output_dim

Is there any way to get around this?

How do you run speculative decoding on vllm with this model?

Can you provide here whole path to run vllm? Preferably from docker.

smcleod

Nov 23, 2024

FYI speculative decoding "just works" with exllamav2 (via TabbyAPI), haven't had any issues using the 1.5b model for the draft.

jukofyork

Nov 28, 2024

How to run speculative decoding of this model with 0.5B model.

I get errors in vllm.

First is that vocab_size is different. So in 0.5B in config.json I set vocab_size=152064 to match with 32B model value.

But now I get error
assert loaded_weight.shape[output_dim] == self.org_vocab_size

I don't see possibility to change output_dim

Is there any way to get around this?

How do you run speculative decoding on vllm with this model?

Can you provide here whole path to run vllm? Preferably from docker.

Just looking at this too, and noticed the 7b model has the same vocab size, but the 0.5b, 1.5b and 3b are all different?

Hrithik2212

Dec 8, 2024

There is a thread on vllm github addressing to this issue
https://github.com/vllm-project/vllm/issues/10913

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