At present, I use vicuna-33b with 3 cards V100 32G memory, and the speed is very slow compared with the demo demo on the official website (https://chat.lmsys.org/). Can you please provide resources for the demo demo on the official website? configuration? Is this speed difference the difference between V100 and A100? Or is it the performance degradation caused by the data interaction of multiple graphics cards in the case of a single machine with multiple cards?
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 39.5k
- Forks
- 4.8k
- PR merge metrics
- No merged PRs in 30d
Description
At present, I use vicuna-33b with 3 cards V100 32G memory, and the speed is very slow compared with the demo demo on the official website (https://chat.lmsys.org/). Can you please provide resources for the demo demo on the official website? configuration? Is this speed difference the difference between V100 and A100? Or is it the performance degradation caused by the data interaction of multiple graphics cards in the case of a single machine with multiple cards?
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 files, tests, or entry points. Start by reviewing the official demo at https://chat.lmsys.org/ and the Vicuna-33B serving configuration, then compare the reported three-V100 setup with the demo hardware and multi-GPU arrangement; done means documenting the relevant configuration and the likely source of the speed difference.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning, performance
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100