vllm-project / vllm-project/aibrix
How to run AIBrix kvcache offload on npu，like 910B
Open
area/kv-cache
kind/documentation
- Dominant language
- Go
- Stars
- 5.1k
- Forks
- 694
- Avg merge
- 1d 19h
- Merged PRs (30d)
- 98
Description
### 🚀 Feature Description and Motivation
image: aibrix-container-registry-cn-beijing.cr.volces.com/aibrix/vllm-openai-aibrix-kvcache:v0.9.1-20250724 can only run on nvidia GPU，now，i want to run on npu,like 910B. how can i do that? which need vllm-ascend.
### Use Case
dist kvcache is very valuable in large model infer，but i don't know how to run it on npu.
### Proposed Solution
_No response_
Contributor guide
Assessment
This issue has not been assessed yet.