Add vLLM as a third inference engine - #9

Open

cguldogan wants to merge 6 commits into

NVIDIA:mainfrom

cguldogan:feat/vllm-engine

Commits on Sep 4, 2026

engine-manager: add the vLLM manifest and a served-model setting

cguldogan
and
claude
committed
2846596
Route vLLM through the OpenAI-compatible proxy

cguldogan
and
claude
committed
6c7ef4d
desktop: surface vLLM as a third engine

cguldogan
and
claude
committed
3f6ec5b
Document vLLM and bump the components it changed

cguldogan
and
claude
committed
4a49cc9
desktop: show "Model to serve" on the local node only

cguldogan
and
claude
committed
215e344
Document vLLM in the remaining hand-maintained docs

cguldogan
and
claude
committed
a4b96a3