huseyinkeles

I have isponsorblocktv running on my home server that skips YouTube segments automatically.

Also running side-loaded Stremio with torrenting enabled so I can stream Big Buck Bunny.

I’ve been using an Apple TV since ~2014 I think. Upgraded to the 4K version on release (2022 version).

Really nothing beats it, especially if you are in Apple ecosystem but even if you are not, it’s a million times better than the messy android tv with full of ads and bloat.

That being said, I don’t see a reason for us to upgrade to this version.

Wow, those are I think the best looking pelicans ever so far

I just used their `Bonsai-demo` repo like this;

`cd ~/Code/Bonsai-demo && BONSAI_CTX=65536 ./scripts/start_llama_server.sh`

then used it in a very minimalistic pi with a very small system prompt.

Didn't spend much time to try to optimize it tbh, but my issue was not the speed. it just could not make a decision on how to implement the task, kept going on an on.

I tried this one, but I found Ornith1.5 to be a better MoE model for me, also very fast. But I still couldn't make it implement a real task on a real repo :( it only worked with an extremely clear directions and very small tasks.

Testing on a MBP m4 pro 24gb

~100t/s prefill, ~15t/s, dropping to ~10t/s later with 64k context.

The issue is I have yet to find a useful agentic local llm that I can run on this machine.

Just given a relatively simple task on a swift app, took 25 minutes, brainstorming like crazy but can not decide on what to do. Eventually I killed it. GPT 5.6 sol-medium took 3 minutes to complete the same task for reference.

when you shoot someone with a gun, who gets sued? you? or the gun manufacturer?

what's next? suing Chrome because it logs in to amazon on my behalf after i give it my amazon credentials?

I spent so much time looking at it, how do they see a leopard there?

I heard this many times before, same thing with the pigs as well, which are another very smart species of animal.

Based on the same logic, is it ethical to eat a human if they are mentally disabled or not passing some IQ test?

I don't know why but your post was marked as [dead] for some reason. Just vouched for it.

I've been experimenting with it on a M4 Pro 24G for the last few hours and it's been very promising using 32k context. getting around 30-40 tps

With Qwen3.8 27B I could not get anywhere near 32k context window, that made it very unusable for agentic coding, although it was very smart.

Related to the #4, I recommend the book Roadside Picnic by Arkady Strugatsky.

“What if we finally got visited by aliens but the most important event in or history was simply an intergalactic roadside picnic for the aliens?

They don’t even bother with us ants. They just pull over have some fun for a couple of hours and leave a mess behind…”

Surely this comment is literally on every new model release post.