Instructions to use OpenMOSS-Team/AnyGPT-base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use OpenMOSS-Team/AnyGPT-base with Transformers:

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

pipe = pipeline("text-generation", model="OpenMOSS-Team/AnyGPT-base")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("OpenMOSS-Team/AnyGPT-base")
model = AutoModelForCausalLM.from_pretrained("OpenMOSS-Team/AnyGPT-base")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use OpenMOSS-Team/AnyGPT-base with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "OpenMOSS-Team/AnyGPT-base"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "OpenMOSS-Team/AnyGPT-base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/OpenMOSS-Team/AnyGPT-base

SGLang

How to use OpenMOSS-Team/AnyGPT-base 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 "OpenMOSS-Team/AnyGPT-base" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "OpenMOSS-Team/AnyGPT-base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "OpenMOSS-Team/AnyGPT-base" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "OpenMOSS-Team/AnyGPT-base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use OpenMOSS-Team/AnyGPT-base with Docker Model Runner:
```
docker model run hf.co/OpenMOSS-Team/AnyGPT-base
```

Improve model card: Add pipeline tag, library name, and detailed inference/training

by nielsr HF Staff - opened Sep 8, 2025

base: refs/heads/main

←

from: refs/pr/1

Discussion Files changed

+95

-31

nielsr

Sep 8, 2025

This PR significantly enhances the model card by adding relevant metadata and comprehensive content from the original GitHub repository.

Key improvements include:

Metadata: Added pipeline_tag: any-to-any to correctly categorize the model and improve discoverability at https://huggingface.co/models?pipeline_tag=any-to-any. Added library_name: transformers based on config.json indicating LlamaForCausalLM architecture and transformers_version, which enables automated usage snippets on the model page.
Structure and Links: Updated the model card title and added prominent badges at the top linking to the paper, project page, and GitHub repository for easy access.
Content Completeness: Integrated the "Open-Source Checklist," "Chat model CLI Inference" instructions, and the detailed "Pretraining and SFT" section from the project's GitHub README, providing a more complete guide for users.
Typos: Corrected the "Lincese" heading to "License".

These updates will improve the model's visibility, usability, and overall presentation on the Hugging Face Hub.

Improve model card: Add pipeline tag, library name, and detailed inference/traininge291f508

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

Tap or paste here to upload images

Ready to merge

This branch is ready to get merged automatically.

· Sign up or log in to comment