Instructions to use BAAI/Emu3.5 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use BAAI/Emu3.5 with Transformers:
# Load model directly from transformers import AutoModelForCausalLM model = AutoModelForCausalLM.from_pretrained("BAAI/Emu3.5", dtype="auto") - Notebooks
- Google Colab
- Kaggle
Add pipeline tag and library name
#1
by nielsr HF Staff - opened
This PR enhances the model card by adding key metadata: pipeline_tag: any-to-any to properly categorize the model's multimodal capabilities, and library_name: transformers to indicate compatibility with the Hugging Face Transformers library for easier usage. The Github link is also added to the model card.
Hey team! It would be really beneficial to merge this PR π€
wolfwjs changed pull request status to merged
can i run this on a single 4090
The model has 34B parameters, so if you quantize that to 4 bits per parameter, then it requires 17.5GB of RAM. Should be feasible on a machine with 24GB of RAM.