Qwen
/

Qwen2.5-72B-Instruct

Text Generation

text-generation-inference

Model card Files Files and versions

Instructions to use Qwen/Qwen2.5-72B-Instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Qwen/Qwen2.5-72B-Instruct with Transformers:

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

pipe = pipeline("text-generation", model="Qwen/Qwen2.5-72B-Instruct")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen2.5-72B-Instruct")
model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2.5-72B-Instruct", 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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
HuggingChat
Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use Qwen/Qwen2.5-72B-Instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Qwen/Qwen2.5-72B-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/Qwen2.5-72B-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Qwen/Qwen2.5-72B-Instruct

How to use Qwen/Qwen2.5-72B-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/Qwen2.5-72B-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/Qwen2.5-72B-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/Qwen2.5-72B-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/Qwen2.5-72B-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Qwen/Qwen2.5-72B-Instruct with Docker Model Runner:
```
docker model run hf.co/Qwen/Qwen2.5-72B-Instruct
```

Resources

View closed (13)

123

#43 opened 4 months ago by

Listed on OpenModelMap

#42 opened 4 months ago by

OMS Score -- OMS 81.9 (A) -- Qwen2.5 72B | 5-dimension independent scoring

#41 opened 5 months ago by

OMS Score -- OMS 81.9 (A) -- Qwen2.5 72B | 5-dimension independent scoring

#40 opened 5 months ago by

OMS Score -- OMS 81.9 (A) -- Qwen2.5 72B | 5-dimension independent scoring

#39 opened 5 months ago by

TemporalMesh Transformer: 29.4 PPL at 48% compute — beats Mamba, new open-source architecture

#38 opened 5 months ago by

System Prompt Bleed causing Knowledge Cutoff Hallucination & Alignment Capitulation

#35 opened 7 months ago by

add AIBOM

#34 opened over 1 year ago by

Inconsistent Output: First API call differs from subsequent identical calls with temperature=0 on Qwen models

#33 opened over 1 year ago by

Rename README.md to levante o tema da evolução criadora de henri bergson

#32 opened over 1 year ago by

ceciliapicinato

Improve language tag

#31 opened over 1 year ago by

Add link to paper

#29 opened almost 2 years ago by

Inference API Body Structure

#28 opened almost 2 years ago by

Model only responding with out of place numbers or letters

#25 opened almost 2 years ago by

model is too busy

#23 opened almost 2 years ago by

Adding Evaluation Results

#21 opened about 2 years ago by

Adding Evaluation Results

#20 opened about 2 years ago by

Adding Evaluation Results

#19 opened about 2 years ago by

Word Count Precision issues with Qwen/Qwen2.5-72B-Instruct: Managing Overlength Responses

#17 opened about 2 years ago by

VenkateshNestor

Default RAG Template?

#16 opened about 2 years ago by

Adding Evaluation Results

#12 opened about 2 years ago by

leaderboard-pr-bot

Adding Evaluation Results

#11 opened about 2 years ago by

leaderboard-pr-bot

Adding Evaluation Results

#10 opened about 2 years ago by

leaderboard-pr-bot

Adding Evaluation Results

#9 opened about 2 years ago by

leaderboard-pr-bot

Adding Evaluation Results

#8 opened about 2 years ago by

leaderboard-pr-bot

Adding Evaluation Results

#7 opened about 2 years ago by

leaderboard-pr-bot

Something weird at layer79.MLP

#6 opened about 2 years ago by

Adding Evaluation Results

#5 opened about 2 years ago by

leaderboard-pr-bot

Does Qwen 2.5 support Thai language?

#4 opened about 2 years ago by

serious censorship

#3 opened about 2 years ago by

Model thinks it's Claude

#2 opened about 2 years ago by

There's a HUGE drop in popular knowledge from v2 to v2.5.

#1 opened about 2 years ago by