Skip to content

Dodaios is an open-source agent OS for your desktop: it runs your coding agents, browser and apps by voice or chat, on your own machine. Formerly Personal Jarvis.

docs: the local-models check is four round trips, and names what it skipped

rubenluetke10-beep552b1a0ebb09
diff --git a/CHANGELOG.md b/CHANGELOG.md
index bed46735df..3ebd24391d 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -81,10 +81,12 @@ the change that makes this 2.0 rather than 1.7.
can do the job it is being considered for. Nothing hides behind the Advanced
switch any more.
- **Local models: the set-up is proven, then saved for next time.** "Set up
- everything" ends with real round trips — the server answers, the chat pick
- answers, the voice round trip completes — shown one line each with how long
- they took (`POST …/local-models/verify`, also the sidebar badge's new
- source), and then switches "Start with Jarvis" on. The Server tab carries
+ everything" ends with four real round trips — the server answers, the chat
+ pick produces an answer, the voice pick calls a tool, the tools & screen pick
+ reads an image — shown one line each with how long they took
+ (`POST …/local-models/verify`, also the sidebar badge's new source). A role
+ that is not configured reports as not run rather than as a pass. It then
+ switches "Start with Jarvis" on. The Server tab carries
both: the switch, with the backend's own sentence on what the next start
would do, and "Run a check". From a terminal:
`jarvis local-models server verify` and `… server autostart on|off`.
diff --git a/README.md b/README.md
index d2a276d21f..7789d5ab18 100644
--- a/README.md
+++ b/README.md
@@ -128,11 +128,13 @@ can take over: chat, voice, tools and screen, and the deep coding work. One butt
four. It starts the server if it is stopped, prefers whatever you already have on disk over
a fresh download, fetches only what is genuinely missing, and writes the settings.
-Then it checks its own work. Three real round trips run at the end, one line each with how
-long it took, so you find out the setup works before you rely on it instead of after. It
-finishes by offering to start the server with Jarvis, which keeps the first answer of the
-day from paying the model load time. Nothing downloads without a click that names the
-download first.
+Then it checks its own work. Four real round trips run at the end: the server answers, the
+chat pick produces an answer, the voice pick calls a tool, and the tools and screen pick
+reads an image. Each reports on its own line with how long it took, and a job you never
+configured says so rather than quietly counting as a pass. So you find out the setup works
+before you rely on it instead of after. It finishes by offering to start the server with
+Jarvis, which keeps the first answer of the day from paying the model load time. Nothing
+downloads without a click that names the download first.
<p align="center">
<img src="https://github.com/PersonalJarvis/PersonalJarvis/raw/main/assets/screenshots/app-local-models.webp" alt="The Local models section: the graphics-memory budget across all four jobs, and a card per job naming the model that fills it" width="900" />