Engineering at Overmatch Prev co-founded Limbic Software (100M+ downloads)
Personal 👉
Joined March 2009
- Tweets2K
- Following429
- Followers1.2K
- Likes1.1K
Super slick new web-based game by @NimbleBit just launched. 🛩️
pilots.nimblebit.com/
Iman Mostafavi retweeted
Very happy to have our new Ice Cream Jump update out on iOS, it's been a long road, too long. Need to finish up the Android version now.
apps.apple.com/us/app/ice-cr…
I think I got at least 6 people to try @spawn today. It is really something special. Amazing work @joexrogan @jsnnsa …this is going to be huge.
I get the pull from 1 Spark up to 4 DGX Sparks. There seem to be a lot of recipes and Local AI community support.
Folks with 4, is there as much utility to adding a 5th?
A DGX Spark has joined my Asus Ascents at home, enabling TP=3. It was barely used from a nice guy on Craigslist for $4,500. He sold it to order a Mac Studio. A decent second hand deal for post Sparkpocalypse times.
It’s a noticeable but not huge speed bump on the current @MiaAI_lab @plotarmordev recipe, but I suspect their upcoming announced recipe powered by @ashxhart TensorFold engine will be even more noticeable. Can’t wait!
This is one of the most impressive research results in AI driven animation to date.
🪄 Bring your 3D characters to life with words!
Introducing UniMate: one unified model for text-driven animation across diverse skeletons—from humans and animals to articulated objects.
🎬 A rigged asset + a text prompt → motion.
No per-skeleton retraining needed!
🌐 linzhanmou.com/unimate/
I snagged @MiaAI_lab's book yesterday and can strongly recommend it for anyone struggling with growing their X account. I've already learned a few critical flaws in my strategy that I can't wait to start correcting. It's no-fluff with actionable information.
Direct link to the book: whop.com/miaai/products/from…
With exactly 4 Sparks are switchless recipes a viable solution or is the overhead of that kind of unofficial cluster setup eating enough of the performance that you really should pony up for the switch?
What’s a useful thing to do with a standalone single Spark that can’t yet join the cluster (waiting on cables)? Throw MiniMax on it?
Used these graphs to convince my work to secure 2 sparks for special projects at the Dallas Micro Center. Thanks for tracking this @loktar00.
This is phenomenal for a locally generated music video. My anthem for today.
TOKENS BURN 🎵
By @NeoAIForecast
Music video generated on a Single Nvidia DGX Spark
with MiniMax H3
Thanks to all who responded to my question about which model is currently best for a small team (2-4) sharing a 2x DGX Spark cluster.
There was consensus around these two models:
🥇Qwen 3.8 Flash Next (recommended)
🥈DeepSeek V4 Flash (faster, more concurrency)
If you're solo, step up to GLM 5.3 Flash. But it's too slow for sharing.