v0.22.1rc1: [docker] Stop using extra-index-url for flashinfer-jit-cache (#44366)
vLLM has released version 0.22.1rc1, which includes a change to stop using extra-index-url for flashinfer-jit-cache. This update addresses a specific technical detail within the project's caching mechanism. The release notes indicate that user feedback is taken seriously and incorporated into development. AI
IMPACT Minor update to an inference engine, unlikely to have broad industry impact.