Instructions to use wangpan-ustc/AtlasVA with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use wangpan-ustc/AtlasVA with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="wangpan-ustc/AtlasVA")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("wangpan-ustc/AtlasVA", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use wangpan-ustc/AtlasVA with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "wangpan-ustc/AtlasVA" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "wangpan-ustc/AtlasVA", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/wangpan-ustc/AtlasVA
- SGLang
How to use wangpan-ustc/AtlasVA 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 "wangpan-ustc/AtlasVA" \ --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": "wangpan-ustc/AtlasVA", "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 "wangpan-ustc/AtlasVA" \ --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": "wangpan-ustc/AtlasVA", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use wangpan-ustc/AtlasVA with Docker Model Runner:
docker model run hf.co/wangpan-ustc/AtlasVA
Improve model card
#1
by nielsr HF Staff - opened
This PR improves the model card by adding relevant metadata and links to the paper, code, and project page.
- Added
library_name: transformers - Added
pipeline_tag: image-text-to-text - Linked the paper: AtlasVA: Self-Evolving Visual Skill Memory for Teacher-Free VLM Agents
- Linked the GitHub repository: https://github.com/wangpan-ustc/AtlasVA
- Linked the project page: https://wangpan-ustc.github.io/AtlasvaWeb/
- Added the base model reference.
wangpan-ustc changed pull request status to merged