Building Transformers.js at 🤗 @huggingface // @GoogleDevExpert for #WebTechnologies and #AI
Thun, Switzerland
Joined April 2015
- Tweets3.5K
- Following1.4K
- Followers3.6K
- Likes5.4K
Huuuge today, cute in 12 months 🤗 Open-source AI is just getting started. Let's go 🚀
We just crossed:
- 20M AI builders on @huggingface
- 300PB of models, datasets, agent traces, and RL environments hosted on the @huggingface hub (9PB added by users just last week)
Also October was our biggest month ever in terms of DAUs, MAUs and revenue.
Let's go open-source AI!
2
301
WebGPU is not *just* a browser feature. WebGPU can also power your React Native app.
And the performance is 🚀
Real-time, local conversational AI is how we will democratize intelligence.
We just ported LFM3-Omni-2B by @LiquidAI to run fully offline on my iPhone 18 Pro Max at ~80 tok/s, powered by custom WebGPU kernels in a React Native app.
Should we release this on the App Store? 👀
Had a great time talking at the #WebAI Summit 🚀
Talks are always more fun when you can share numbers like this: Transformers.js went from 1.64M monthly npm downloads at last year's summit to ~15M today. 9x in a year.
Thanks to everyone building with it 🤗
2
17
682
Speaking at @codetalkshh in Hamburg next week!
Nov 4, 15:00, main stage: “Boring but Effective: WebAI That Ships”
ML that isn’t a chatbot: embeddings, speech recognition, background removal. All in the browser with Transformers.js + WebGPU, fully offline
1
11
406
👉 codetalks.com/agenda?view=ti…
As a little gift you get 50% off your ticket with CT50-FINALCALL, valid until the event with no usage limit.
1
146
Also, the evening before (Nov 3) there will be the Agentic Coding Meetup at Google Hamburg. I’ll show you how to go from raw recording to finished video in 15 min by just prompting @Remotion. luma.com/agentic-dco0
1
86
Exciting week ahead: flying to the Bay Area on Wednesday for WebAI Summit! It's a short trip, just three days, so if you want to grab a coffee, drop me a DM 😊
1
10
721
I turned structured output into a game 🎮
The LLM scores the next tokens. You pick one. The constraints decide if it’s allowed.
Most likely ≠ valid. Even a capital letter can lose you points.
Watch the demo. Play it yourself below 👇
3
14
485
Your turn: can you get 100/100?
Try heads or tails, then build a tiny JSON creature one token at a time.
@googlegemma 4 E2B runs locally in your browser via Transformers.js + WebGPU.
Play Token Arcade:
huggingface.co/spaces/nico-m…
1
84
New York, here I come! 🗽
On Nov 16 I'm speaking at JSNation US about multimodal AI agents that see, hear and remember, all running entirely in your browser.
Voice, vision and memory, all running in the browser. No backend, no API keys 🙌
@nicodotdev, Machine Learning Engineer at @huggingface, shows how WebGPU and WebAssembly make real-time AI agents possible on the open web.
Join us at JSNation US 👇
jsnation.us/
2
12
314
For all the german speaking people, I'll be on the token therapie podcast with @janschultecom and Michael Köpf live in about 5 minutes: youtube.com/watch?v=Ixx05pt1…
9
635
More Ternary Bonsai 2 models from @PrismML just dropped: 2B, 4B, 9B, 27B 🌱
Hundreds of tok/s on consumer hardware, fully in the browser with custom WebGPU kernels. At ~1.8 bits/weight the 2B is just 430MB. Local agentic AI is getting wild 🤯
Today has been one of the biggest days for open-source AI this year... and it keeps getting better.
@PrismML just released more Ternary Bonsai 2 models on Hugging Face: now in 2B, 4B, 9B, and 27B. The 2B reaches up to 500 tok/s on my M4 Max ENTIRELY in the browser with custom WebGPU kernels... that's insane! 🤯
At ~1.8 bits per weight, the language-model weights are just 430MB for 2B and 950MB for 4B (plus ~365MB for the 8-bit vision weights). The future of local agentic AI is going to be wild.
Try out the demo yourself 👇
Next week I'll be at F{AI}R, an AI unconference in Liebefeld, Bern 🇨🇭
The best part of unconferences: the schedule is built together on the spot, and the best moments usually happen over coffee, not on stage.
4
11
532
I'll bring some input on running AI directly in the browser, but mostly I’m looking forward to the conversations.
Got a topic? Submit a session. Entry is free.
fair.siliconlovefield.ch/sch…
2
89
Your turn: can you get 100/100?
Try heads or tails, then build a tiny JSON creature one token at a time.
@googlegemma 4 runs locally in your browser via Transformers.js + WebGPU. First visit downloads the model.
Play Token Arcade:
huggingface.co/spaces/nico-m…
2
106
The prompt says what we want. The constraints define what’s allowed.
Token Arcade lets you step through that distinction yourself: probabilities, token choices, rejected continuations, and explanations.
Code:
github.com/nico-martin/token…
44