Model Downloads¶
IdeaWeaver provides a simple and efficient way to download models from the Hugging Face Hub. The download functionality allows you to:
- Download models directly from Hugging Face Hub
- Specify model variants and versions
- Automatically handle model dependencies
- Verify downloaded model integrity
Basic Usage¶
The simplest way to download a model is using the download
command:
ideaweaver download <model_name>
For example, to download the DialoGPT-medium model:
ideaweaver download microsoft/DialoGPT-medium
Output Example¶
When downloading a model, you'll see output similar to this:
🔍 Downloading model: microsoft/DialoGPT-medium
📦 Model files:
- config.json
- pytorch_model.bin
- tokenizer.json
- tokenizer_config.json
- special_tokens_map.json
✅ Model downloaded successfully to: ./models/microsoft/DialoGPT-medium
Model Directory Structure¶
After downloading, the model will be stored in a directory structure like this:
./models/
└── microsoft/
└── DialoGPT-medium/
├── config.json
├── pytorch_model.bin
├── tokenizer.json
├── tokenizer_config.json
└── special_tokens_map.json
Next Steps¶
- Learn about download commands for advanced options
- Check out the fine-tuning guide to start training with your downloaded model
- Explore RAG capabilities to use your model in a retrieval-augmented generation system