py3.10-vllm-cuda-11.8
Chainguard
Status
Impact
This issue only affects the V0 engine, which has been off by default since v0.8.0. Further, the issue only applies to a deployment using tensor parallelism across multiple hosts, which upstream does not expect to be a common deployment pattern. Since V0 has been off by default since v0.8.0 and the fix is fairly invasive, Upstream has decided not to fix this issue. Instead recommend that users ensure their environment is on a secure network in case this pattern is in use.
Status