Loading video...

Video Failed to Load

Go Home

Meta's Segment Anything Model (SAM) can now run in your browser w/ WebGPU (+ fp16), meaning up to 8x faster image encoding (10s → 1.25s)! 🤯⚡️ Video is not sped up! Everything runs 100% locally thanks to 🤗 Transformers.js and onnxruntime-web! 🔗 Demo:

xenovacom's profile picture

Xenova

17,017 subscribers

120,352 views • 2 years ago •via X (Twitter)

Science & TechnologyNews & Politics
LIVE

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

Run OpenAI's new Whisper Turbo model 100% locally in your browser with Transformers.js! ⚡️ Transcribe 2 minutes of audio in ~12 seconds! 🤯 Demo + source code 👇
0:15

Run OpenAI's new Whisper Turbo model 100% locally in your browser with Transformers.js! ⚡️ Transcribe 2 minutes of audio in ~12 seconds! 🤯 Demo + source code 👇

Xenova

137,002 views • 1 year ago

IBM just released Granite 4.0 1B Speech, a compact and efficient speech-language model, designed for multilingual speech recognition and bidirectional speech translation. New #1 on the OpenASR leaderboard! It can even run in your browser on WebGPU, thanks to 🤗 Transformers.js
0:14

IBM just released Granite 4.0 1B Speech, a compact and efficient speech-language model, designed for multilingual speech recognition and bidirectional speech translation. New #1 on the OpenASR leaderboard! It can even run in your browser on WebGPU, thanks to 🤗 Transformers.js

Xenova

21,506 views • 3 months ago

WOW! 🤯 Language models are becoming smaller and more capable than ever! Here's SmolLM2 running 100% locally in-browser w/ WebGPU on a 6-year-old GPU. Look at that speed! ⚡️😍 Powered by 🤗 Transformers.js and ONNX Runtime Web! How many tokens/second do you get? Let me know! 👇
0:13

WOW! 🤯 Language models are becoming smaller and more capable than ever! Here's SmolLM2 running 100% locally in-browser w/ WebGPU on a 6-year-old GPU. Look at that speed! ⚡️😍 Powered by 🤗 Transformers.js and ONNX Runtime Web! How many tokens/second do you get? Let me know! 👇

Xenova

12,557 views • 1 year ago

Opus 4.7 just wrote a custom WebGPU kernel that runs Qwen3.5 up to 13x faster using a fused LinearAttention op! 🤯 Agentic kernel optimization is the future. Now live in 🤗 Transformers.js v4.2.0! P.S. I've updated all our previous demos to use this new version. Enjoy!
0:10

Opus 4.7 just wrote a custom WebGPU kernel that runs Qwen3.5 up to 13x faster using a fused LinearAttention op! 🤯 Agentic kernel optimization is the future. Now live in 🤗 Transformers.js v4.2.0! P.S. I've updated all our previous demos to use this new version. Enjoy!

Xenova

78,136 views • 2 months ago

🟢 News: GPU compute in the browser is finally real. WebGPU now ships by default in Chrome, Firefox, Safari, and Edge—not a polyfill, not behind a flag. You can run LLMs client-side. Transformers.js and ONNX Runtime already ship WebGPU backends. Eight years of spec work. No more asterisks. 🔗 #WebGPU #GPU #Chrome #Safari #Firefox #Edge
0:10

🟢 News: GPU compute in the browser is finally real. WebGPU now ships by default in Chrome, Firefox, Safari, and Edge—not a polyfill, not behind a flag. You can run LLMs client-side. Transformers.js and ONNX Runtime already ship WebGPU backends. Eight years of spec work. No more asterisks. 🔗 #WebGPU #GPU #Chrome #Safari #Firefox #Edge

WebGL / WebGPU

12,762 views • 5 months ago

Stable Diffusion 100% in the browser with running at < 1s per image ⚡️ Thanks to SD Turbo by Axel Sauer & al Stability AI and diffusers.js on webGPU by Arthur Islamov Demo: More info in ↓🧵
0:13

Stable Diffusion 100% in the browser with running at < 1s per image ⚡️ Thanks to SD Turbo by Axel Sauer & al Stability AI and diffusers.js on webGPU by Arthur Islamov Demo: More info in ↓🧵

Cyril Diagne

96,600 views • 2 years ago

There has been a huge debate recently about the best approach for image background removal. Here's my attempt: - In-browser inference w/ 🤗 Transformers.js - WebGPU accelerated (fast!) - Costs $0 (no image hosting or server processing) - No data leaves your device (privacy!)
0:15

There has been a huge debate recently about the best approach for image background removal. Here's my attempt: - In-browser inference w/ 🤗 Transformers.js - WebGPU accelerated (fast!) - Costs $0 (no image hosting or server processing) - No data leaves your device (privacy!)

Xenova

419,020 views • 1 year ago

yapped so good they sped him up 8x faster to make it quick
0:15

yapped so good they sped him up 8x faster to make it quick

koi

80,523 views • 4 months ago

Introducing 🤗 Transformers.js v4: state-of-the-art machine learning for the web! 🚀 New WebGPU backend (browser, Node.js, Bun, Deno) ⚡️ Huge performance improvements 🤯 Support for over 200 architectures 🛠️ Complete codebase refactor Learn more about our biggest release yet! 👇
0:12

Introducing 🤗 Transformers.js v4: state-of-the-art machine learning for the web! 🚀 New WebGPU backend (browser, Node.js, Bun, Deno) ⚡️ Huge performance improvements 🤯 Support for over 200 architectures 🛠️ Complete codebase refactor Learn more about our biggest release yet! 👇

Xenova

72,710 views • 3 months ago

Let’s fucking goo!! DeepSeek R1 1.5B running FULLY LOCALLY in your browser at 60 tok/ sec powered by WebGPU🔥 Intelligence truly is too cheap to meter! ⚡️
0:13

Let’s fucking goo!! DeepSeek R1 1.5B running FULLY LOCALLY in your browser at 60 tok/ sec powered by WebGPU🔥 Intelligence truly is too cheap to meter! ⚡️

Vaibhav (VB) Srivastav

973,002 views • 1 year ago

Real-time SmolVLM in a web-browser with transformers.js. All running locally with no installs. Just open the website.
0:17

Real-time SmolVLM in a web-browser with transformers.js. All running locally with no installs. Just open the website.

Andi Marafioti

86,299 views • 1 year ago

290MB = useful LLM (running directly in your browser via WebGPU) 🤯 Bonsai 1-bit is beyond my comprehension... try the demo on hugging face ⬇️
0:10

290MB = useful LLM (running directly in your browser via WebGPU) 🤯 Bonsai 1-bit is beyond my comprehension... try the demo on hugging face ⬇️

Victor M

63,007 views • 2 months ago

Anything now runs full browser tests agents open your app click every button run every flow QA, fully automated
0:18

Sensitive content

Anything now runs full browser tests agents open your app click every button run every flow QA, fully automated

Anything

23,129 views • 8 months ago

Today we're releasing the Segment Anything Model (SAM) — a step toward the first foundation model for image segmentation. SAM is capable of one-click segmentation of any object from any photo or video + zero-shot transfer to other segmentation tasks ➡️
0:15

Today we're releasing the Segment Anything Model (SAM) — a step toward the first foundation model for image segmentation. SAM is capable of one-click segmentation of any object from any photo or video + zero-shot transfer to other segmentation tasks ➡️

AI at Meta

3,570,378 views • 3 years ago

Did you know we're working with environmental survey teams to improve flood detection? Our Segment Anything Model (SAM) is being used to identify minute changes in water conditions from satellite images—we hope this faster analysis can contribute to better rapid response and keep communities safer.
0:16

Did you know we're working with environmental survey teams to improve flood detection? Our Segment Anything Model (SAM) is being used to identify minute changes in water conditions from satellite images—we hope this faster analysis can contribute to better rapid response and keep communities safer.

Meta

157,982 views • 4 months ago

2025 was a landmark year for local machine learning. We're looking back at how Transformers.js and WebGPU enabled state-of-the-art models right in the browser. See how to build the future of the web with @HuggingFace's Xenova →
0:17

2025 was a landmark year for local machine learning. We're looking back at how Transformers.js and WebGPU enabled state-of-the-art models right in the browser. See how to build the future of the web with @HuggingFace's Xenova →

Chrome for Developers

16,753 views • 6 months ago

Cursor and Windsurf both launched new models. So I decided to put them to the test in speed, up to date info, and app creation. Here are the full breakdown & results: (Video is not sped up 🤯)
0:16

Cursor and Windsurf both launched new models. So I decided to put them to the test in speed, up to date info, and app creation. Here are the full breakdown & results: (Video is not sped up 🤯)

Alvaro Cintas

83,712 views • 8 months ago

WOW! 😍 Sakana AI just released TinySwallow, a 1.5B Japanese LLM created through TAID (Temporally Adaptive Interpolated Distillation), a new knowledge distillation technique. 🐦 hardmaru even spoke about in-browser use on Bloomberg, "we show that [TinySwallow] works entirely inside of a smartphone or entirely inside your web browser, without it calling an API." Love it! 🔥 I recreated their demo with 🤗 Transformers.js, and it's up to 2x faster (50-60 tokens per second on an M3 Max)! Check it out! 👇
0:18

WOW! 😍 Sakana AI just released TinySwallow, a 1.5B Japanese LLM created through TAID (Temporally Adaptive Interpolated Distillation), a new knowledge distillation technique. 🐦 hardmaru even spoke about in-browser use on Bloomberg, "we show that [TinySwallow] works entirely inside of a smartphone or entirely inside your web browser, without it calling an API." Love it! 🔥 I recreated their demo with 🤗 Transformers.js, and it's up to 2x faster (50-60 tokens per second on an M3 Max)! Check it out! 👇

Xenova

17,269 views • 1 year ago

What I love about Replit ⠕. Low floor to get started. Ultra high ceiling for what you can build. I told Agent to build a native C application that runs on my desktop without a web browser - AND compile it into JavaScript to run in the web browser. Here's a video showing both running
0:15

What I love about Replit ⠕. Low floor to get started. Ultra high ceiling for what you can build. I told Agent to build a native C application that runs on my desktop without a web browser - AND compile it into JavaScript to run in the web browser. Here's a video showing both running

jordwalke

43,739 views • 4 months ago

feeling good with load times now :) (video is not sped up)
0:18

feeling good with load times now :) (video is not sped up)

alan

16,942 views • 1 month ago