MAT retweeted
Replying to @JamesTervit
That is awesome, thank you!
I just started documenting this method in terms of skills for agents. The install should be super easy.
Would love your take on it!
github.com/gmkurtzer/hdae
MAT retweeted
Got an old Mac in a drawer? A 2017 MacBook Pro is running Omarchy here and it's flying. Old hardware isn't done, it just needed the right OS.
MAT retweeted
Three weeks ago @NotJustShaking had never touched Linux. Now Lee is making Wired Differently with the Omarchy crowd behind him. Go give it a watch and tell him what you think.
MAT retweeted
Omarchy question for the weekend: what's the first thing you'd tell a brand new user to change? Best answers go in a beginners' list I'm putting together.
MAT retweeted
Replying to @JamesTervit @ncfrontiersman
The Radeon does the work: its hardware video decoder sat around 20% playing 1080p HEVC straight from the file. No cloud, no transcoding, no subscription. The one limit: AV1, which this 2017 GPU can't decode.
MAT retweeted
Everything you see came across over wayshare: Omarchy's screen streamed straight into the Mac's own Screen Sharing app on my M3, smooth enough that we recorded every clip right off the share. Screen-sharing with @ncfrontiersman. Narrated by a local voice clone.
MAT retweeted
This 2017 MacBook Pro (T1, Radeon Pro) is running Omarchy and Uber TV, an open-source media center.
An AI agent found my Plex server and synced 590 titles in 21 seconds. Then it played them on a GPU most people wrote off 🧵📺
MAT retweeted
Next for the Mule itself: its orchestrator moves to DeepSeek, running MiaAI-Lab's GB10 build on our two DGX Spark-class GB10 systems.
Once testing is done, in a few weeks, Claude, Codex and Grok retire from our pipeline. Orchestration, render and voice, all on hardware we run.
MAT retweeted
What consistency looks like: the same pack mule at dawn, at the server rack and in the handover, and every list playing over moving footage, timed to the voice.
12 min 48 s. 74 shots. 105 narrated lines. Rendered on two NVIDIA GB10 boxes, cut on a Mac.
MAT retweeted
On 12 October we could make about one minute of consistent AI film. Today it is 12 minutes.
Same mule in every shot. One narrator. Real-time motion. Made locally, on our own GPUs, with open models.
Next: 30 minutes. Then three 30-minute segments for 90, and on to 120.
MAT retweeted
Replying to @JamesTervit @Artz2023
How we work with early customers:
- Samples first, for your approval
- Your characters and voices kept consistent (consent needed for any real person)
- Finals rendered and delivered to you
- Early-partner pricing
MAT retweeted
Replying to @ncfrontiersman
@ncfrontiersman one more wayshare PR: a new mute_local option mutes the Linux machine's speakers while a Screen Sharing viewer gets the audio, so the screenshare host becomes the only output. Unmutes on disconnect and host gets sound back on again.
github.com/wesleygrimes/ways…
MAT retweeted
The full ModelMule film, 12:48.
Every picture, voice line and note came from local models on our own GPUs and Macs. No cloud, no subscription.
What ModelMule is, how members back serious compute, and where it goes next.
Please give us a repost to help us bring out the big guns.
modelmule.com
@TheModelMule @ncfrontiersman @NixFred @NotJustShaking @thetundeo @calledtocnstrct @finbar_b @Smeekraft
MAT retweeted
Hey @X! I’m at 843 followers, aiming high to hit 3k 🚀
I’m building ModelMule, a non-profit IRC hub for open-weight local AI models, with a simple interface for people new to IRC. Keep Local AI Alive 🔥
Next up: bringing a GB300 online for compute-constrained folks doing serious AI development. Looking for 600 backers at $1k each to launch with unmetered access ⚡
Help me #connect 🤝 with people who get it:
🦙 Uncensored weights > API keys
🔥 Your GPU is too small and you know it
📡 IRC never died, it was waiting for us
🐧 Omarchy or nothing
🛠️ Building real infra in public
Follower count doesn’t matter, good conversations do! ✨
ko-fi.com/modelmule
MAT retweeted
Replying to @JamesTervit @home_assistant
Update: we did it for real. @MDH54H2's GivEnergy system is back under his control, fully local.
His app stopped working after the administration. The inverter kept running; he just couldn't see or control it. Now it talks straight to Home Assistant on his own network.
MAT retweeted
The nostalgia of my local llm recreating all my old emulator games. having fun loading all my old code from yesteryear...
















