Choose a runtime from product constraints

The native vLLM path currently depends on a community-converted checkpoint; use the official FunASR split-engine when an official-weight chain is required. Evidence covers one H100 only, so retest the target GPU, audio, languages, hotwords, and traffic.

A TensorRT plan is not portable across arbitrary GPU architectures or TensorRT versions; rebuild and revalidate accuracy, memory, and capacity on the target stack.

Prebuilt GPU packages cover only the documented backend and architecture; the Windows AMD Vulkan crash fix still awaits reporter hardware confirmation, and every other target also requires hardware validation.

Official verification currently covers Linux source builds; the server does not include production authentication, TLS, tenant quotas, or durable queues, so provide them at the gateway and infrastructure layers.

SenseVoice is merged into audio.cpp main@979e070f but is not yet in a tagged release. Pin that main commit until a release is published. Neither path currently provides word-level timestamps.

The example server does not include production authentication, tenant quotas, or full observability; provide them at the gateway and infrastructure layers.

Capacity depends on realtime ratio, chunks, VAD, connection count, and client backpressure; file RTFx does not replace concurrency tests.

The CPU example is not a CUDA production image; GPU drivers, image, scheduling, and capacity still require platform validation.

Model conversion, operator support, and thread settings vary by platform; measure capacity on the target CPU with real audio.

A checklist does not replace workload, accuracy, failure, and security exercises for the target product.