Instructions to use facebook/llm-compiler-13b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use facebook/llm-compiler-13b with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("facebook/llm-compiler-13b", dtype="auto") - Notebooks
- Google Colab
- Kaggle
A problem encountered by the llm compiler when generating IR from source code
#2
by terry2334 - opened
I use the prompt shown below to let the 13b-ftd model output IR from the source code
Then I entered the source code as follows
and encountered the output without IR as follows
This problem doesn't seem to occur in the large model. It should output the corresponding IR. Can anyone explain why this phenomenon occurs?