Skip to main content

Activity for emsh.cat

Active on:

Loading activity...

Posted: Huh, this might not be entirely wrongBluesky network
Posted: Another hand/finger tracking experiment, this time using the built-in monochrome cameras on the Steam Frame! Everything runs on-device, inc…Bluesky network
Posted: I'll do the best... Boris Mann?1!?Bluesky network
PostedBluesky network

I love it when things are just fast and simple, why over-complicate things? Self-hosted #forgejo is just cherry on the top

PostedBluesky network

I guess I'm officially a professional #VRDev now then, thanks @itch.io! \o/

Posted: Ah fuck, it's another triangle, isn't it? It's always a triangle somehow, and "chose two *huew huewh huweh*"Bluesky network
PostedBluesky network

MatineeVR - Open Source Native and hardware-accelerated VR video playback on Steam Frame (Experimental / WIP) embedding-shapes.itch.io/matineevr #Steam #SteamOS #SteamFrame #FOSS #OSS #VR

Posted: I found no native Steam Frame applications for local playback standalone on the device itself, so I made it myself! :) Native ARM64 binar…Bluesky network
Posted: Almost embarrassed to say this but, I just got the Steam Frame and one of the first things I felt like I had to do once inside the desktop …Bluesky network
Posted: Whats the most important today isn't what you chose to build, its what you don't chose to buildBluesky network
Posted: Spent a day or two with Sol Max attempting to recover the files from Btrfs, seems 64/68 files are restorable! All may not be lost yet.Bluesky network
PostedBluesky network

Trying something different, a more personal post about AI, authenticity, club nights and why a room full of strangers can feel more human than anything on a screen. Kind of out there for me, but here goes nothing :) emsh.cat/en/the-peopl... #AI #MusicSky #ElectronicMusic #WritingCommunity

The People Are Still There
emsh.cat
Posted: Was building a harness for Qwen3.8-27B by using Qwen3.8-27B locally, to see how useful it is for real work. I was keeping all the session …Bluesky network
PostedBluesky network

Alright, the Codex Micro just arrived! Now lets see how I can use this not with Codex but with my own harness, and hopefully also forcefully add a microphone to this device :D

  • Photo of the Codex Micro
PostedBluesky network

It's all downhill from here, hold on to your horses

Posted: When our corporate overlords play opposite-day for a conference, what a fucking jokeBluesky network
Posted: It seems like #OpenAI is planning to encrypt all the prompts that #codex uses on disk, we can already not see whatever the main agent sends…Bluesky network
PostedBluesky network

Microsoft displaying the prowess of vibe-coding yet again, is there seriously no QA left at MS?

PostedBluesky network

It really is! Just achieved ~400 tok/s in single prompt text-generation! Text-only still, but still :) #DiffusionGemma feels like it has a really good foundation for local AI. Loading a 50GB model into VRAM + 1024 tokens inference in under 8 seconds in total, I think this changes everything :)

  • Stats: 1024 tokens in 2.39s (429.00 tok/s), 43 diffusion steps (18.01 it/s)
env CUDA_HOME=/opt/cuda CUDA_PATH=/opt/cuda CUDA_ROOT=/opt/cuda         cargo  5,00s user 2,41s system 99% cpu 7,454 total