Roland 1dd33d26e2 ai: keep slow-CPU chat streams alive ... On the FX-8320 the first token took over a minute, and Sanic's default 60-second RESPONSE_TIMEOUT killed the stream before anything was sent. Raise the timeout above the chat loop's own 600s hard ceiling, and emit a status event before every model call so the stream is never silent — for proxies and for the user staring at a slow machine. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>		2026-09-13 20:10:17 -06:00
..
ai	ai: keep slow-CPU chat streams alive	2026-09-13 20:10:17 -06:00
archive	search: don't COUNT(*) OVER() the page query (#702)	2026-09-12 14:44:30 -06:00
docs	docs: browse walks the file_group date index instead of probing every doc (#707)	2026-09-13 12:39:05 -06:00
flasher	Flasher: chip-mismatch guard, WROLPi firmware table, colorblind chip colors	2026-07-02 17:26:37 -06:00
inventory	Strip whitespace from inventory item string values on save	2026-07-30 17:34:46 -06:00
map	Map previews that never drew, and a Docker map with no world under it (#658)	2026-08-11 20:37:23 -06:00
videos	search: don't COUNT(*) OVER() the page query (#702)	2026-09-12 14:44:30 -06:00
zim	zim: recompute the outdated-Zims flag after startup and on Delete (#706)	2026-09-12 15:19:36 -06:00
__init__.py	Move One-Time Pad encryption to the browser (#479)	2026-06-13 13:52:34 -06:00