Open Models Almanac
Methodssoftware implementationinference engine

single mention · software implementation

vLLM tensor parallelism across eight GPUs

A model-serving configuration that uses vLLM with tensor parallel size eight.

The sources on this page only mention it in passing. The line above is the extraction's one-line gloss. See inference engine for the methods filed beside it.

Also called vLLM at --tensor-parallel-size 8.

Filed under software implementation › inference engine