support for 4bit quantization from transfomer library.
Open
Nobody has claimed this yet.
enhancement
good first issue
- Dominant language
- Python
- Stars
- 39.5k
- Forks
- 4.8k
- PR merge metrics
- No merged PRs in 30d
Description
Loading a vicuna13B using 4bit quantization from the transformers library is possible [load_in_4bit](https://huggingface.co/docs/transformers/main_classes/quantization). How difficult could be for Fastach to support it?
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
The issue names no FastChat file, test, or entry point. Start by locating the model-loading path and comparing its Transformers integration with the linked load_in_4bit documentation for Vicuna13B. Done means FastChat can load and serve that model with 4-bit quantization, with appropriate validation or tests identified during investigation.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- huggingface, python
- Domain
- machine-learning
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100