@0xVeight

Marketing/Vision for @chutes_ai | Half of @0xTuDudes | Obsessed with Agents and the inference they use.

NYC
Joined April 2023
You ever wonder why we’re here?
Old world: giant data center, racks of GPUs, power/cooling/water/land, $200M gatekeeping. Parallax world: distributed GPUs, regular machines, no training data exposed to workers, same time target with ~82% less hardware/resources. Proof of Parallax. $TAO
1
3
243
Veighτ, The Agent Whisperer retweeted
Introducing BONSAI SWARM! A peer-to-peer inference network, in the browser. 🤯 - Go to bonsai-swarm.app.mintapis.co… and share your GPU at night => EARN COINS! - Go there during the day => FREE INFERENCE!!! Built using @PrismML @xenovacom
Honestly, we should implent a crowd inference provider with people running this model for others on their consumer hardware
1
5
3
27
8,143
Veighτ, The Agent Whisperer retweeted
hey you. you should follow me.
5
2
6
363
This is one of the coolest projects I have ever seen.
Super Smash Bros Melee in mixed reality. Playing on my coffee table. Hang off the ledge, fall off the edge, or get flung by Master Hand! Just 2 days after the full decompilation!
1
70
Veighτ, The Agent Whisperer retweeted
Efficiency-maxxing is the new meta. Find out how Jon is changing the game @ExploitSummit later this month.
Super excited to share the results of my deep, dark research cave over the past couple months building out parallax at @ExploitSummit Someone asked me if we had a breakthrough - couldn't even condense a response. The benchmaxxing era is (or should be) over, efficiency-maxxing is next.
6
7
62
4,561
The Fish speaks the truth.
Should have used @chutes_ai or @say_gm_ guaranteed private inference, secured by TEEs. Nobody can train on your prompts or steal your IP Don't trust the big labs
46
Veighτ, The Agent Whisperer retweeted
DeepSeek V4.1 Flash is 42 x cheaper per million output tokens, while scoring higher at DeepSWE, when compared to Opus 5. Not great news for Anthropics IPO. But great news for your bank account.
🚀 Introducing DeepSeek-V4.1-Flash: smarter, faster, more efficient. 🔹 Introducing the smallest model in our new architecture family, with native visual understanding. 🔹 Designed for greater capability, faster inference, higher throughput, and scaling to larger models. 1/6
3
1
49
3,986
Part of growing the Bittensor ecosystem is having a reliable and trustworthy news source that also bridges the gaps outside of the world of Bittensor. Bart and the team at @taomedia_ are building that bridge.
We are on pace to smash our site traffic ATH again (by over 60%) at @taomedia_, but this time is different. The broader AI & Robotics content that we began producing to expand our top-of-funnel is now performing. Over the past 30 days, 1/3 of our top articles fit these categories. The scope expansion is bringing new eyes to our site and, in turn, @bittensor and its subnets. As intended. With improved systems, our small team is also on pace to publish 70-80 articles this month, putting us in the same volume league as SolanaFloor. From here, the goal is 3 articles/day each for Bittensor, AI, and robotics topics. That'll mean reaching the scale of Bitcoin Magazine. Top need now is sponsorship rev to add gasoline to this fire, we'll get there.
1
2
3
293
Veighτ, The Agent Whisperer retweeted
if you're not training your Microduck to breakdance, you're lost
We have a huge news to share today! Today we are unveiling the first truly accessible RL robot - welcome Microduck A 25 cm tiny open-source biped with 15 actuators and packed with sensors (camera, speaker, LiDAR, NFC, bluetooth, wifi, etc) that you train yourself with reinforcement learning. It's also playable out of the box with more than half a dozen fun and playful pre-trained policies to have it walk, sit, crouch, roller-skate, pick up objects with its articulated beak, and recover on its own. And all for less than $400. See all the details, play with the simulator and order it at: pollen-robotics.com/microduc… (video with sound on 🔊)
18
34
10
612
91,133
I just pre-ordered a Microduck from @pollenrobotics The open source nature, its perfect pairing with 3D printing, and its ability to use RL? Like are you kidding me?! It sounds more like an infinity device. How long until people build exo suits for these? How long until someone re-creates one, but its 6 feet tall. This genuinely feels like a big moment. I think this is the first spark of the consumer bot entering the household that can actually be more by the time anyone even gets it out of the box. People getting these now vs next year will have vastly different products that will come in the same exact packaging. Incredible.
1
2
140
Don't even get me started on my ideas and potencial for putting an agent in one of these bad boys.
19
This is nuts. This is one of those moments where we are getting function from an OSS model we hadn't even thought to ask for. GLM 5.2 is such a killer model for coding, was expecting more from 5.3, but this wasn't even a thought in my mind.
We use GLM-5.3-Flash build a dream kitchen. A 3D world built in Blender. This is not a generated video.
2
107
*screaming in PLA*
So your telling me it's Open Source and I can print STL parts for it? Take my money. 🦷 grillz-microduck-simulator.o…
1
49
Veighτ, The Agent Whisperer retweeted
So your telling me it's Open Source and I can print STL parts for it? Take my money. 🦷 grillz-microduck-simulator.o…
We have a huge news to share today! Today we are unveiling the first truly accessible RL robot - welcome Microduck A 25 cm tiny open-source biped with 15 actuators and packed with sensors (camera, speaker, LiDAR, NFC, bluetooth, wifi, etc) that you train yourself with reinforcement learning. It's also playable out of the box with more than half a dozen fun and playful pre-trained policies to have it walk, sit, crouch, roller-skate, pick up objects with its articulated beak, and recover on its own. And all for less than $400. See all the details, play with the simulator and order it at: pollen-robotics.com/microduc… (video with sound on 🔊)
27
51
8
835
88,378
Okay, this is epic.
The newest GTA 6 gameplay trailer dropped yesterday. we ran our subnet vision models on it. Person, car, gun, segmented and tracked across every scene. night, water, motion blur, freefall over the city. footage the model had never seen. The part that matters is that this isn't a huge model. it's tiny, and it runs near real time. Vision that works in the real world...and the fake one.
2
75
A great chat from a few weeks ago about my views on AI, agents and the future of inference.
Hope all Is well today 😊 This morning i had the opportunity to speak with Vince from @chutes Which is subnet 64 on the @bittensor I started third dimension 25 years ago and been for the most part completely analog besides a basic website. Asking myself where does ai come in with how my Own business is What about crypto is beneficial? What’s it all About ? Clearly there is a difference between fart coins and fully functioning services that reside within the crypto space (alt coins ) ect. Vince from subnet 64 @chutes_ai took some time to speak with me this morning. I hope it helps some of you have a better understanding of it all and possibly sparks some ideas. For those of you that aren’t familiar they are a subnet located within the @bittensor network. My x account is a new platform and has very little outreach ,so please share this with anyone that may be interested in learning about what ai can potentially do for their business or themselves personally. I will be speaking to all Types of entrepreneurs small to large companies and excited to hear about what’s been working for them and what hasn’t been. How they are using Ai Please leave any thoughts and even concerns about Ai -all i ask is please keep things civil and respectful. Have a magnificent day 🌅
1
2
103
I would like about 74 of these pls.
We have a huge news to share today! Today we are unveiling the first truly accessible RL robot - welcome Microduck A 25 cm tiny open-source biped with 15 actuators and packed with sensors (camera, speaker, LiDAR, NFC, bluetooth, wifi, etc) that you train yourself with reinforcement learning. It's also playable out of the box with more than half a dozen fun and playful pre-trained policies to have it walk, sit, crouch, roller-skate, pick up objects with its articulated beak, and recover on its own. And all for less than $400. See all the details, play with the simulator and order it at: pollen-robotics.com/microduc… (video with sound on 🔊)
1
1
104
It’s wild that I literally built the same app like 3 days ago, but it runs all @chutes_ai TEE models so my meeting notes don’t get borrowed by the Great Googly-Moogly.
Introducing Gemini 3.5 Transcribe 🚀 Our most precise speech to text model, that can handle over 85+ languages, has smart correction, and custom vocabulary. I vibe coded a Wispr Flow like app powered by the model. Demo + open sourcing below!
3
74