From 6804c785c6b506c47b45264398728d0a609310be Mon Sep 17 00:00:00 2001 From: historia Date: Wed, 2 Sep 2026 19:13:24 -0400 Subject: feat: sglang-omni model picker on manual server launch --- app/docs/backend-sglomni.md | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'app/docs/backend-sglomni.md') diff --git a/app/docs/backend-sglomni.md b/app/docs/backend-sglomni.md index 2aac93e..bf7d3a9 100644 --- a/app/docs/backend-sglomni.md +++ b/app/docs/backend-sglomni.md @@ -118,6 +118,13 @@ hosts one model, a run whose selected model differs from the hosted one restarts a server this tool started — a foreign server hosting another model refuses the run with an actionable message instead. +The hub's **Start/Stop Backend Servers** menu asks which *downloaded* model +to load when starting the server fresh (with none downloaded the entry says +so and points at the Configure screen; a single model starts without +asking). A start also auto-installs a model's missing companion packages +first, so a boot never dies on their import even when the weights arrived +via the shared HuggingFace cache or an earlier install's pip run failed. + The server boots a multi-stage pipeline (preprocessing → TTS generation → vocoder) and may pull companion weights on first start, so its start timeout is larger than the other backends' (20 minutes). Pre-downloading -- cgit v1.2.3