
co-founder @VoltaAudio | building @madeira_app | computers, AI, rockets, robotics | i make cool headphones btw
Joined July 2016
- Tweets1.4K
- Following2.6K
- Followers1K
- Likes17.6K
Pinned Tweet
6 months ago, I had a dream of running RDR2 on an iPhone. I wasn't sure if it could be done, but I decided to try anyways, by building a general-purpose Windows PC games emulator for iOS.
I am extremely happy to announce that this dream has come true!
Will Faust retweeted
It's happening: Nintendo Switch 2 emulation/recomps on PC is now a thing.
Incredible how we went from nothing to first emulation results in just a few days.
Will Faust retweeted
CPU working. NES emu is full speed now. Like I said, exciting times! #switch2 #switch2hacks
Wow they finally did the 30B class model! Looks amazing, I’ll start downloading it right away :)
🚀Excited to announce Nanbeige4.3-33B!
To push the performance ceiling of small agents, our strategy in Nanbeige4.3 starts with building a powerful teacher first.
Nanbeige4.3-33B delivers SOTA agentic performance among open models at ~30B parameters:
* Strong results on agentic coding and cowork benchmarks.
* When integrated into mainstream agent harnesses, it tackles real-world tasks with cost-efficient inference.
* It will serve as a teacher for upcoming smaller Nanbeige models, starting with Nanbeige4.3-3B.
huggingface.co/Nanbeige/Nanb…
huggingface.co/Nanbeige/Nanb…
#LLM #Agent
🚀 ¿Quieres jugar tus juegos de STEAM y PC en tu iPhone o iPad? 🎮
📱 En este video te muestro cómo instalar y configurar Madeira para jugar juegos de computadora directamente en iOS, activar JIT desde tu dispositivo, conectar Steam o jugar cualquier juego de PC.
No te lo pierdas aquí 👇
#emulador #iphone #gaming #steam
🔗 youtu.be/4bpZ0DqThsA
This project has been very fun to work on, with much more coming soon :)
30 FPS with perfect rendering in RDR2 should be possible!
Windows games can now run on iOS! Sneak peak of what I'm working on...
Will Faust retweeted
Ox Alpha is frontier. wtf. Seriously wtf. Tencent? Xiaomi? Really? It's more well-done than any other Chinese model. I swear, it's better than Kimi K3. It's good *in ways that Chinese models are typically horrible at*. It's FAST
I can half-believe it's a stolen Claude checkpoint.
42 on DeepSWE at 27B is absolutely insane! I hope it holds up in practice. I'll test it on my MacBook right now, thank you Qwen team!
Gonna need to make an emergency trip for this, at all cost
Replying to @DJSnM
Unless we discover problems after mission data review, SpaceX will attempt to catch the ship with the tower on next flight
Nanbeige4.2-3B technical report is out on @huggingface !
huggingface.co/Nanbeige/Nanb…
We released Nanbeige4.2-3B, its Looped Transformer increases model capacity without adding parameters, delivering a capable 3B agent.
Nanbeige4.5 is training with LoopSplit, mHC+depth attention & concatenated n-gram embeddings,already in the modeling code.
huggingface.co/Nanbeige/Nanb…

















