Instructions to use CognitiveKernel/Qwen3-8B-CK-Pro with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use CognitiveKernel/Qwen3-8B-CK-Pro with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="CognitiveKernel/Qwen3-8B-CK-Pro")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("CognitiveKernel/Qwen3-8B-CK-Pro", dtype="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use CognitiveKernel/Qwen3-8B-CK-Pro with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "CognitiveKernel/Qwen3-8B-CK-Pro" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "CognitiveKernel/Qwen3-8B-CK-Pro", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/CognitiveKernel/Qwen3-8B-CK-Pro
- SGLang
How to use CognitiveKernel/Qwen3-8B-CK-Pro 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 "CognitiveKernel/Qwen3-8B-CK-Pro" \ --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": "CognitiveKernel/Qwen3-8B-CK-Pro", "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 "CognitiveKernel/Qwen3-8B-CK-Pro" \ --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": "CognitiveKernel/Qwen3-8B-CK-Pro", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use CognitiveKernel/Qwen3-8B-CK-Pro with Docker Model Runner:
docker model run hf.co/CognitiveKernel/Qwen3-8B-CK-Pro
Improve model card: Add comprehensive information and usage
#1
by nielsr HF Staff - opened
This PR significantly enhances the model card for the CognitiveKernel/Qwen3-8B-CK-Pro model by adding crucial metadata and comprehensive information:
- Metadata:
- Adds
pipeline_tag: image-text-to-text, which correctly categorizes the model for tasks involving both images and text, improving discoverability on the Hugging Face Hub. - Adds
library_name: transformers, indicating the model's compatibility with the Hugging Face Transformers library.
- Adds
- Paper Link: Includes a direct link to the research paper: Cognitive Kernel-Pro: A Framework for Deep Research Agents and Agent Foundation Models Training.
- Project Page: Adds a link to the official project homepage: https://osatlas.github.io/.
- Code Repository: Provides a direct link to the GitHub repository: https://github.com/OS-Copilot/OS-Atlas.
- Model Description and Usage: Adds a clear overview of the model's capabilities and a practical Python code snippet for inference using the
transformerslibrary, complete with an example image input and question. - Visual Aid: Includes an image from the project's GitHub repository for better visual context.
- Citation: Adds the official BibTeX citation for the paper.
These additions will greatly improve the visibility, usability, and reproducibility of the model for the community.