Valentin

Work on mobile too of course

With resizeable activity when used in desktop mode.

And it’s working in desktop mode too (pixel 11 in desktop mode)

I really like the idea, android and iOS app is almost ready here (with support of resizeable activity) Demo in 4g (really far away from home, via tailscale)

Is this really better? With llama.cpp Ingram and Q4 unsloth I have : ngram-map-k 7 (retenu) │ Vulkan0 │ 301 │ 54,0 (Halo Strix 128gb)

Une relique en vidant une boîte @Ubuntu_fr @Canonical 2010 ça date ! #ubuntu #shipit

Nice , it makes my AI max+ Hermes compatible without too much latency. But in my setup, the reasoning level seems always on regardless of the configuration. Did I miss something?

Nice , it makes my AI max+ Hermes compatible without too much latency. But in my setup, the reasoning level seems always on regardless of the configuration. Did I miss something?

Interesting, in my setup, I have the speed of the 27b (~12t/s Strix Halo 128gb). The draft model seems loaded (visible in the logs)

@YaLTeRz I having an issue with the is-window-cast-target=true match need something specific ? I tried using the sharing feature of Chrome. But my rules are not applied (I use the one from the documentation)