AI @GoogleDeepMind | Author of "Hands-On LLMs" (llm-book.com) & "Illustrated Guided to AI Agents" (ai-agents-book.com) | RAG Pack (ragpack.ai)
- Tweets 391
- Following 38
- Followers 10,289
- Likes 297
Maarten Grootendorst retweeted
Gemini 4 Argon is here.
It’s available globally today for developers and enterprises in the paid tier of the Gemini API and the Gemini platform, for Ultra subscribers in Google Antigravity, and Ultra and Pro subscribers in the Gemini app and Search.
Hope you all enjoy it!
1,547
2,481
28,470
5,928,249
Over 300 original figures, hundreds of iterations, two years of hard work... and it's finally here!
An Illustrated Guide to AI Agents is being shipped and delivered to people all over the world. @JayAlammar and I are excited to share this work with you all and to finally be able to see it for ourselves.
Visuals, concepts, and even code. We have something in the book for everyone!
3
4
30
1,934
Maarten Grootendorst retweeted
Friends in India! The @shroffpub edition is out, at a much friendlier price, in full color, and with free shipping in India.
This is your chance to dig into the 330+ figures @MaartenGr and I crafted to explain the most important concepts in AI agents.
🚀 NOW SHIPPING! 📚🤖
An Illustrated Guide to AI Agents by @MaartenGr & @JayAlammar is here!
Explore LLMs, tools, memory, planning, reasoning, multi-agent systems & practical code.
Order now shroffpublishers.com/books/9…
#AIAgents #AI #LLM #MachineLearning
3
2
20
3,155
Join the RAG Pack for a free masterclass on Agents!
I'm looking forward to this one soooo much! @JayAlammar , @joshuastarmer, @ChrisJMcCormick, @SerranoAcademy, and I worked our asses off to make this one extra special.
Oh, and you might see one or two visuals from "An Illustrated Guide to AI Agents" 😉
2
3
16
1,596
The eBook version of "An Illustrated Guide to AI Agents" is live!
More than 300 illustrations and the code for building an Agent from scratch. @JayAlammar and I did WAY too much for this book, and I'm glad we did.
We learned so much from our previous book, and we did all we could to make this book even more special. You will learn how to build an Agent entirely from scratch through text, code, and figures. We wanted to give you the complete educational experience and focused heavily on a single narrative throughout the entire book.
The physical edition should land in a couple of weeks, and I'm excited to get my hands on it!
Amazon: amazon.com/Illustrated-Guide…
O'Reilly: oreilly.com/library/view/an-…
5
28
253
8,696
Maarten Grootendorst retweeted
yesterday, we released agentic video understanding with Gemini.
here's a great video explainer from @MaartenGr on how it works:
piped.video/ytjgy30Cono
5
6
47
6,284
Maarten Grootendorst retweeted
Ever wonder what makes Gemma 4 12B so unique?
In this 2 minute overview series, learn what an encoder-free model is and how Gemma 4 12B processes image and audio inputs differently from the rest of the Gemma family!
21
78
841
43,500
Maarten Grootendorst retweeted
The complete DiffusionGemma technical report is officially live!
We're sharing our full process and insights to help the community explore the incredible potential of text diffusion.
We’re excited to see how this approach drives new research and applications.
33
127
1,232
71,337
Maarten Grootendorst retweeted
still the best book on AI agents I’ve picked up in 2026
from the great @MaartenGr and @JayAlammar.
buy it here:
→ amazon.com/Illustrated-Guide…
you can start reading it for free here :)
→ oreilly.com/library/view/an-…
19
114
738
37,355
Join us for another masterclass tomorrow!
It's about Reinforcement Learning with super fun sessions by @SerranoAcademy , @joshuastarmer , @JayAlammar , @ChrisJMcCormick , and myself!
1
2
13
1,122
Introducing Gemma 4 Drafters!
Up to a 3x speed up with guaranteed same quality for all model checkpoints (E2B, E4B, 31B, 26B A4B) 🚀
2
14
102
5,029
I couldn't help myself and added MANY visuals to explain how they work: newsletter.maartengrootendor…
1
3
18
1,292
This was a fun demo to make 😀
What does it take to run 3, 5, or even 10 concurrent instances of Gemma 4 locally?
We've open-sourced a demo letting you run multiple models side-by-side on your hardware.
Gemma 4 26B A4B easily runs 10+ concurrent requests on a MacBook Pro M4 Max at 18 tokens/sec per request.
3
1
32
3,639
Maarten Grootendorst retweeted
If only every foundational LLM lab would publish such a guide with their model releases.
A Visual Guide to Gemma 4
With almost 40 (!) custom visuals, explore the new models from Google DeepMind. We explore various techniques, ranging from Mixture of Experts and the Vision Encoder all the way up to Per-Layer Embeddings and the Audio Encoder.
Link below 👇
7
70
619
67,302