Instructions to use MegaScience/Qwen3-8B-MegaScience with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use MegaScience/Qwen3-8B-MegaScience with Transformers:

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

pipe = pipeline("text-generation", model="MegaScience/Qwen3-8B-MegaScience")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("MegaScience/Qwen3-8B-MegaScience")
model = AutoModelForCausalLM.from_pretrained("MegaScience/Qwen3-8B-MegaScience")
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]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use MegaScience/Qwen3-8B-MegaScience with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "MegaScience/Qwen3-8B-MegaScience"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "MegaScience/Qwen3-8B-MegaScience",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/MegaScience/Qwen3-8B-MegaScience

SGLang

How to use MegaScience/Qwen3-8B-MegaScience 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 "MegaScience/Qwen3-8B-MegaScience" \
    --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": "MegaScience/Qwen3-8B-MegaScience",
		"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 "MegaScience/Qwen3-8B-MegaScience" \
        --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": "MegaScience/Qwen3-8B-MegaScience",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use MegaScience/Qwen3-8B-MegaScience with Docker Model Runner:
```
docker model run hf.co/MegaScience/Qwen3-8B-MegaScience
```

Improve model card: Add metadata, abstract, GitHub link, and usage example

by nielsr HF Staff - opened Jul 24, 2025

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

+61

-10

nielsr

Jul 24, 2025

This PR enhances the model card for MegaScience/Qwen3-8B-MegaScience by:

Adding library_name: transformers to the metadata, enabling the "How to use" widget.
Adding relevant tags for better discoverability.
Including the paper's abstract for a comprehensive overview.
Providing a direct link to the official GitHub repository.
Adding a Python usage example to facilitate model loading and inference.

Improve model card: Add metadata, abstract, GitHub link, and usage examplee87dae3b

Vfrz

MegaScience org Jul 24, 2025

Thank you very much for your effort in refining the README.

Vfrz changed pull request status to merged Jul 24, 2025

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