Local LLM inference · GPU homelabs & DGX · ES/EN
Madrid, Spain
Joined September 2009
- Tweets2.8K
- Following818
- Followers3.6K
- Likes6.2K
Pinned Tweet
2x DGX Spark join the family on top of my rtx pro 6000, 5090's and 3090's. The plan for this is to run 24/7 with my agents as the low power usage is really appealing.
I'm going to start with @MiaAI_lab DSv4 recipe and might also try @gerdelgado repo
Impressive little machines @NVIDIAAI
day 1 at railway. pushed a tiny fix to env var storage. status page looking spicy but i think i did good
God of War III running at 120fps on Linux. What a time time to be alive.
Heroic launcher and GE Proton 11-7 work perfectly on Arch Linux and a 5090
Btw there is also another one for GLM-5.4
opencode.ai/en/data/zhipu/gl…
But errors out on 5.6 or other numbers
Is this a placeholder or are you leaking new releases @opencode @thdxr @Zai_org
opencode.ai/en/data/zhipu/gl…
Is this a placeholder or are you leaking new releases @opencode @thdxr @Zai_org
opencode.ai/en/data/zhipu/gl…
Drop the magnet link anon
We discovered a sophisticated recording device that was hidden inside one of our offices. We used our models to understand the device and where it had come from, identifying that a specific employee had planted it a few days earlier. We referred the matter to the FBI, and we’re grateful to the agents and prosecutors who moved quickly to make an arrest. The FBI’s investigation is ongoing and we will continue to support their work.
At this time, we have no indications of this individual inappropriately accessing or exfiltrating model weights, source code, or customer data. The employee was terminated.
We take protecting our technology, data, and people seriously, and as a matter of national importance. Our security program is designed to protect against a wide range of threat actors, from insider risks to cybercriminals to nation states, and this case is an example of how AI can help address these sophisticated security threats.
Rodri Mora aka Bullerwins retweeted
Me and @MiaAI_lab talked it out and came to an understanding. There was a misunderstanding on both ends, and we cleared it up.
The bottom line: nobody wins when people in the local AI community, all working to push open source forward, are fighting each other. There's no side to pick here. We're all chasing the same goal, and fighting only slows the process down.
Appreciate @SeraAndroid for helping make the conversation happen. Back to building 🤝
I'm glad to share that @Tech2Wild and I talked it through and came to a mutual understanding that lets us both focus on what actually matters:
Pushing local AI forward.
That's been the real goal for both of us from the start. I truly believe the open source and local AI community is better off working together, not against each other.
Special thanks to @SeraAndroid for bridging the gap between us and helping make this conversation happen.
Now, back to work!
Rodri Mora aka Bullerwins retweeted
I'm glad to share that @Tech2Wild and I talked it through and came to a mutual understanding that lets us both focus on what actually matters:
Pushing local AI forward.
That's been the real goal for both of us from the start. I truly believe the open source and local AI community is better off working together, not against each other.
Special thanks to @SeraAndroid for bridging the gap between us and helping make this conversation happen.
Now, back to work!
If I know something is that the second @alexandr_wang trims the mullet Muse is over
I was able to replicate this locally with @MiniMax_AI H3 and a FirstFrame-LastFrame workflow
This is the workflow I used civitai.com/models/2901812/m…
The DGX Spark at home is a niche in the LocalLLM world, which is itself a niche in the agentic AI world, which is itself a niche in the broad AI/ChatGPT world that everyone knows.
Drama in this niche of a niche of a niche was not in my 2026 bingo card.
Rodri Mora aka Bullerwins retweeted
I tried to decompile the Fitbit APK, perform reverse engineering in order to not depend on Google's services to use my Fitbit watch.
At this point, it has been impossible; furthermore, some actions, such as performing a factory reset, are server-side (wtf)
Can someone help me?
