Skip to content

Conversation

22dimensions
Copy link
Contributor

@22dimensions 22dimensions commented Feb 22, 2025

Motivation

see issue #3781

Modifications

  1. add new device args 'npu'
  2. import torch_npu and adapt torch related code
  3. add usage document

Checklist

Examples

python -m sglang.launch_server --model-path LLM-Research/Llama-3.2-1B-Instruct  --device npu --grammar-backend outlines --attention-backend torch_native  --port 30000 --host 0.0.0.0

image

@22dimensions 22dimensions changed the title [WIP][Feautre, Hardware] add suport for ascend npu [WIP][Feautre, Hardware] add initial suport for ascend npu Feb 22, 2025
@22dimensions
Copy link
Contributor Author

#3853 is merged, so i close this pr.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant