This website requires JavaScript.
Explore
Help
Sign in
andodeki
/
makepad
Watch
1
Star
0
Fork
You've already forked makepad
0
Code
Issues
Pull requests
Projects
Releases
Packages
Wiki
Activity
Actions
work
makepad
/
libs
/
ai
/
hub
History
Download ZIP
Download TAR.GZ
Exact
Exact
Union
RegExp
Admin
a45c57d534
ai: CUDA kernels behind a cuda feature that only the AI Hub enables (apps stay CPU-only; no CUDA rebuild ping-pong between hub and app launches)
...
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-26 09:47:27 +01:00
..
examples
ai: the AI hub gives way to the UI, smaller models, vision and speech
2026-10-13 12:26:21 +02:00
prompts
ai: the AI hub gives way to the UI, smaller models, vision and speech
2026-10-13 12:26:21 +02:00
python
ai: the AI hub gives way to the UI, smaller models, vision and speech
2026-10-13 12:26:21 +02:00
src
Desk Link: rate creeps only while pictures flow (to 1.25x measured throughput), new sessions start at the viewer's last rate or 250 Mbit/s, logged on change; AI hub on a wildcard front recognises its own machine (no endless local-use 403)
2026-10-26 05:21:44 +01:00
tests
work: AI Hub desktop app (in-process hub, Enabled + per-service switches, idle turn-off, fleet status, shared hub panel) + cross-platform tray API (macOS menu bar, Windows notification area)
2026-10-25 13:51:31 +01:00
Cargo.toml
ai: CUDA kernels behind a cuda feature that only the AI Hub enables (apps stay CPU-only; no CUDA rebuild ping-pong between hub and app launches)
2026-10-26 09:47:27 +01:00
registry.json
work: AI hub model prefetch (every runnable model on every node from fleet peers, licence-aware), registry pins + sha256, peer secret in ai-fleet
2026-10-25 03:20:08 +01:00