Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

Simulating a fruit fly's neural network using custom WebGPU kernels, running locally in a browser. The demo uses real connectivity data for 166,700 neurons from the MaleCNS connectome. Paint neurons and watch their activity drive the fly's movements. Built with GPT-6 Astra.

Xenova

20,669 subscribers

125,684 views • 1 day ago •via X (Twitter)

Science & Technology Gaming Health & Wellness

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

A Neural Network Can Grow New Neurons Where It Is Confused? In 1994, Bernd Fritzke published A Growing Neural Gas Network Learns Topologies. He introduced a network that starts small, follows incoming data, and inserts new neurons where its error is highest. In the animation, the fog is the drifting data. The glowing nodes are neurons. The fibers are learned connections. The network grows into a living skeleton of the manifold.

A Neural Network Can Grow New Neurons Where It Is Confused? In 1994, Bernd Fritzke published A Growing Neural Gas Network Learns Topologies. He introduced a network that starts small, follows incoming data, and inserts new neurons where its error is highest. In the animation, the fog is the drifting data. The glowing nodes are neurons. The fibers are learned connections. The network grows into a living skeleton of the manifold.

Mathelirium

38,306 views • 3 months ago

Bonsai 27B just changed the local LLM game forever. 1-bit quantization shrinks it from 54GB to just 3.8GB (-93%), while retaining 90% of its intelligence. That's insane. With custom WebGPU kernels written by Fable 5 and GPT 5.6 Sol, the model now runs locally in your browser!

Bonsai 27B just changed the local LLM game forever. 1-bit quantization shrinks it from 54GB to just 3.8GB (-93%), while retaining 90% of its intelligence. That's insane. With custom WebGPU kernels written by Fable 5 and GPT 5.6 Sol, the model now runs locally in your browser!

Xenova

211,051 views • 1 month ago

It's official: Qwen3.8-27B just scored 52 on the Artificial Analysis Intelligence Index. We now have an open-weight model that matches GPT-5.6 Luna (max) AND can run locally... even in your browser with custom WebGPU kernels! What a time to be alive! 🤯

It's official: Qwen3.8-27B just scored 52 on the Artificial Analysis Intelligence Index. We now have an open-weight model that matches GPT-5.6 Luna (max) AND can run locally... even in your browser with custom WebGPU kernels! What a time to be alive! 🤯

Xenova

186,674 views • 21 days ago

someone built a BIOLOGICAL AI HYBRID, real living neurons on a multi-electrode CL1 array that influence how Grok THINKS in real time its called bioLLM you send a message, it gets encoded into electrical signals and delivered to a living neural culture running on CL1 by Cortical Labs, the neurons fire, they measure spike activity across channels and compute a consciousness score that score becomes Grok temperature dial, if neurons firing hard AI gets creative, unpredictable, bold word choices, when culture resting, it precise, conservative, focused the creator is using Grok specifically because xAI API gave the most introspective outputs, says Neuralink + xAI resources make it the right long-term bet

someone built a BIOLOGICAL AI HYBRID, real living neurons on a multi-electrode CL1 array that influence how Grok THINKS in real time its called bioLLM you send a message, it gets encoded into electrical signals and delivered to a living neural culture running on CL1 by Cortical Labs, the neurons fire, they measure spike activity across channels and compute a consciousness score that score becomes Grok temperature dial, if neurons firing hard AI gets creative, unpredictable, bold word choices, when culture resting, it precise, conservative, focused the creator is using Grok specifically because xAI API gave the most introspective outputs, says Neuralink + xAI resources make it the right long-term bet

chiefofautism

71,554 views • 6 months ago

NVIDIA AI Aerial uses neural networks to learn from data, adapt in real time, and scale with complexity. This boosts spectral efficiency, lowers costs, and improves connectivity for telecom. To learn more, watch the full video ➡️

NVIDIA AI Aerial uses neural networks to learn from data, adapt in real time, and scale with complexity. This boosts spectral efficiency, lowers costs, and improves connectivity for telecom. To learn more, watch the full video ➡️

NVIDIA AI

11,275 views • 1 year ago

Using monkey brain data to operate robotic arms! 🙉 Project Jenkins explores brain-computer interfaces by decoding neural activity into robotic movement and generating synthetic brain data from motion patterns. Using neural recordings from a macaque monkey named Jenkins, researchers developed models to decode brain signals into robotic arm movements and encode movement data into simulated neural activity. The project utilizes robotic arms and an interactive web console for real-time brain data generation from joystick inputs. The research paves the way for adaptive brain-computer interfaces by modeling brain activity more flexibly. Congrats to authors Dima Yanovsky, and Andrii Zahorodnii! 🦾

Using monkey brain data to operate robotic arms! 🙉 Project Jenkins explores brain-computer interfaces by decoding neural activity into robotic movement and generating synthetic brain data from motion patterns. Using neural recordings from a macaque monkey named Jenkins, researchers developed models to decode brain signals into robotic arm movements and encode movement data into simulated neural activity. The project utilizes robotic arms and an interactive web console for real-time brain data generation from joystick inputs. The research paves the way for adaptive brain-computer interfaces by modeling brain activity more flexibly. Congrats to authors Dima Yanovsky, and Andrii Zahorodnii! 🦾

Lukas Ziegler

21,627 views • 1 year ago

GPT-6 Astra is the strongest model for 3D game from a single prompt. We described the game in one sentence > GPT-6 Astra wrote the whole thing - drift physics, combo scoring, near-miss bonuses, speed traps, nitro > out came Street Heat, a full arcade racer running in the browser. The whole pipeline ran end-to-end on Higgsfield Supercomputer.

GPT-6 Astra is the strongest model for 3D game from a single prompt. We described the game in one sentence > GPT-6 Astra wrote the whole thing - drift physics, combo scoring, near-miss bonuses, speed traps, nitro > out came Street Heat, a full arcade racer running in the browser. The whole pipeline ran end-to-end on Higgsfield Supercomputer.

Higgsfield AI 🧩

170,655 views • 3 days ago

Introducing Voxtral WebGPU: Real-time speech transcription entirely in your browser. This demo runs Voxtral-Mini-4B, a powerful streaming ASR model from Mistral AI, locally on WebGPU. The model supports 13 languages and is capable of <500 ms latency. Fully private. Zero cost.

Introducing Voxtral WebGPU: Real-time speech transcription entirely in your browser. This demo runs Voxtral-Mini-4B, a powerful streaming ASR model from Mistral AI, locally on WebGPU. The model supports 13 languages and is capable of <500 ms latency. Fully private. Zero cost.

Xenova

94,356 views • 6 months ago

The full video from Cortical Labs explaining how they put 200,000 brain cells onto a silicon chip and had it play Doom is wild: “When a demon appears on the left of the screen, specific electrodes stimulate the sensory area of the neural culture on the left side. The neurons react to that stimulation. We then listen to their response, the spikes, and interpret that activity as motor commands. If the neurons fire in a specific pattern, the Doom guy shoots.”

The full video from Cortical Labs explaining how they put 200,000 brain cells onto a silicon chip and had it play Doom is wild: “When a demon appears on the left of the screen, specific electrodes stimulate the sensory area of the neural culture on the left side. The neurons react to that stimulation. We then listen to their response, the spikes, and interpret that activity as motor commands. If the neurons fire in a specific pattern, the Doom guy shoots.”

Trung Phan

233,872 views • 6 months ago

Before Fable 5 was shut down, it pushed Gemma 4 to 255 tok/s on WebGPU. Some didn't believe it was real. Today we're releasing the demo and kernels it wrote for you to see yourself. Run it locally in your browser. Agentic kernel optimization is the future of on-device inference

Before Fable 5 was shut down, it pushed Gemma 4 to 255 tok/s on WebGPU. Some didn't believe it was real. Today we're releasing the demo and kernels it wrote for you to see yourself. Run it locally in your browser. Agentic kernel optimization is the future of on-device inference

Xenova

484,504 views • 2 months ago

built a weird experiment with GPT 6 Astra so there's a person sitting at a computer running a local model, you talk to him, he turns around and listens, then turns back to the PC and uses ai for you you can literally build anything now

built a weird experiment with GPT 6 Astra so there's a person sitting at a computer running a local model, you talk to him, he turns around and listens, then turns back to the PC and uses ai for you you can literally build anything now

Leon Lin

64,246 views • 2 days ago

can a neural network learn to walk as a physical object in a physics simulation? here I train walking neural nets with an evolutionary algorithm. The input nodes/feet are activated by sine waves at learned phases & connections between two neurons extend based on their difference

can a neural network learn to walk as a physical object in a physics simulation? here I train walking neural nets with an evolutionary algorithm. The input nodes/feet are activated by sine waves at learned phases & connections between two neurons extend based on their difference

Matt Henderson

339,340 views • 3 years ago

RF-DETR, the state-of-the-art model series for real-time object detection, can now run 100% locally in your browser on WebGPU with 🤗 Transformers.js v4! The models are Apache-2.0 licensed, making them a perfect fit for both personal and commercial applications. Try the demo 👇

RF-DETR, the state-of-the-art model series for real-time object detection, can now run 100% locally in your browser on WebGPU with 🤗 Transformers.js v4! The models are Apache-2.0 licensed, making them a perfect fit for both personal and commercial applications. Try the demo 👇

Xenova

77,179 views • 6 months ago

I used GPT-6 Astra to get AAA PC games running on my iPad mini! Skyrim, Batman: Arkham City, Hades, and Age of Empires II. The actual Windows versions, running locally with touch controls. No streaming. Demo below and I'll share a repo and setup guides soon!

I used GPT-6 Astra to get AAA PC games running on my iPad mini! Skyrim, Batman: Arkham City, Hades, and Age of Empires II. The actual Windows versions, running locally with touch controls. No streaming. Demo below and I'll share a repo and setup guides soon!

Ammaar Reshi

279,539 views • 12 hours ago

There's a fruit fly walking around right now that was never born. eon just released a video where they took a real fly's connectome — the wiring diagram of its brain — and simulated it. Dropped it into a virtual body. It started walking. Grooming. Feeding. Doing what flies do. Nobody taught it to walk. No training data, no gradient descent toward fly-like behavior. This is the opposite of how AI works. They rebuilt the mind from the inside, neuron by neuron, and behavior just... emerged. It's the first time a biological organism has been recreated not by modeling what it does, but by modeling what it is. A human brain is 6 OOM more neurons. That's a scaling problem, something we've gotten very good at solving. So what happens when we have a working copy of the human mind?

There's a fruit fly walking around right now that was never born. eon just released a video where they took a real fly's connectome — the wiring diagram of its brain — and simulated it. Dropped it into a virtual body. It started walking. Grooming. Feeding. Doing what flies do. Nobody taught it to walk. No training data, no gradient descent toward fly-like behavior. This is the opposite of how AI works. They rebuilt the mind from the inside, neuron by neuron, and behavior just... emerged. It's the first time a biological organism has been recreated not by modeling what it does, but by modeling what it is. A human brain is 6 OOM more neurons. That's a scaling problem, something we've gotten very good at solving. So what happens when we have a working copy of the human mind?

hattie

9,571,560 views • 6 months ago

Late night experiment. I built a demo that generates 3D environments from images into 3DGS, then uses them as a spatial reference for directing real-time AR camera movements. The capture is then run through Seedance 2.0 for the final output. all running on the web via three.js↓

Late night experiment. I built a demo that generates 3D environments from images into 3DGS, then uses them as a spatial reference for directing real-time AR camera movements. The capture is then run through Seedance 2.0 for the final output. all running on the web via three.js↓

Ian Curtis

28,168 views • 2 months ago

To highlight the new v2 API integrations for x402, Ash built a demo app on Replit ⠕ showing how teams can build agents that can autonomously retrieve data from the web using a wallet with stablecoin payments. Watch the demo 👇

To highlight the new v2 API integrations for x402, Ash built a demo app on Replit ⠕ showing how teams can build agents that can autonomously retrieve data from the web using a wallet with stablecoin payments. Watch the demo 👇

Coinbase Developer Platform🛡️

16,290 views • 11 months ago

⚡Technical Showcase: Peer-to-peer GPU sharing, entirely in the browser. AI Grid lets you run LLMs locally with WebGPU and either donate your spare compute to the network or borrow from someone else's machine. No installs, no cloud. Decentralized inference with nothing but a URL. 🔗 #WebGPU #WebLLM #WebAI #DistributedCompute #ArtificialInteligence

⚡Technical Showcase: Peer-to-peer GPU sharing, entirely in the browser. AI Grid lets you run LLMs locally with WebGPU and either donate your spare compute to the network or borrow from someone else's machine. No installs, no cloud. Decentralized inference with nothing but a URL. 🔗 #WebGPU #WebLLM #WebAI #DistributedCompute #ArtificialInteligence

WebGL / WebGPU

17,200 views • 7 months ago

Discover how to add AI-driven features to your site AND keep user data local using Chrome built-in AI APIs. We partnered with Nykaa Fashion to show how an AI model running inside the browser processes images to suggest personalized outfits. Watch the demo showing a dynamic experience with on-device AI →

Discover how to add AI-driven features to your site AND keep user data local using Chrome built-in AI APIs. We partnered with Nykaa Fashion to show how an AI model running inside the browser processes images to suggest personalized outfits. Watch the demo showing a dynamic experience with on-device AI →

Chrome for Developers

13,574 views • 1 month ago