Package
lmcache-cuda-12.8
Component
vllm
Latest update
6.5
CVSS V3
Build, ship, and run secure software with minimal, hardened container images — rebuilt from source daily and guarded under our industry-leading remediation SLA.
Start for freeStatus
Impact
Fixed upstream in vLLM 0.30.0. This package pins vllm 0.19.1, the last release installable on a CUDA 12.8 stack: every vLLM release from 0.20.0 onward pins torch 2.11.0 (0.27.0+ torch 2.13.0), whose PyPI wheels require CUDA 13 bindings, so resolution against cuda-bindings~=12.8 is unsatisfiable. The affected structured-output token validation, which passes speculative-decoding padding tokens to the grammar backend, is present in 0.19.1. Remediated in lmcache-cuda-13.0.
Status