Instructions to use BlackSamorez/llama-13b-hf-tensor-parallel with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use BlackSamorez/llama-13b-hf-tensor-parallel with Transformers:
# Load model directly from transformers import AutoTokenizer, AutoModel tokenizer = AutoTokenizer.from_pretrained("BlackSamorez/llama-13b-hf-tensor-parallel") model = AutoModel.from_pretrained("BlackSamorez/llama-13b-hf-tensor-parallel") - Notebooks
- Google Colab
- Kaggle
Example of how to use
#1
by milyiyo - opened
Hi @BlackSamorez :)
Do you have any Colab showing how to use these models or a Space in HuggingFace?
Hi @milyiyo !
It's a model for tensor_parallel library and it' supposed to be ran on 2 GPUs
Here's the demo: https://www.kaggle.com/code/blacksamorez/tensor-parallel-llama-13b/
@BlackSamorez How do you convert a llama hf model into a tensor parallel model? Thanks. Asking since I am trying to test the 30b model.
@diegomontoya
It's detailed in the project's readme.