Poka
Poka
When will this feature be available? Thanks.
Hi team, Has this issue been resolved? I couldn't find it in v0.3.2. #822 If there are any considerations, please feel free to share them. Thanks~
@maobaolong Thank you~ It seems to be related to sparse attention, which adds an indexer, thereby doubling the kv array size (`kv_cache_pointers`). [https://blog.vllm.ai/2025/09/29/deepseek-v3-2.html](url)
@maobaolong Thank for the solution! Yesterday, I attempted to build lmcache on an ARM64 (aarch64) environment using the vLLM nightly image, but it failed. I’d also like to ask if...