LLM Architecture Gallery(sebastianraschka.com)
345 points bytzury14 hours ago |22 comments
libraryofbabel6 hours ago
This is great - always worth reading anything from Sebastian. I would also highly recommend his Build an LLM From Scratch book. I feel like I didn’t really understand the transformer mechanism until I worked through that book.

On the LLM Architecture Gallery, it’s interesting to see the variations between models, but I think the 30,000ft view of this is that in the last seven years since GPT-2 there have been a lot of improvements to LLM architecture but no fundamental innovations in that area. The best open weight models today still look a lot like GPT-2 if you zoom out: it’s a bunch of attention layers and feed forward layers stacked up.

Another way of putting this is that astonishing improvements in capabilities of LLMs that we’ve seen over the last 7 years have come mostly from scaling up and, critically, from new training methods like RLVR, which is responsible for coding agents going from barely working to amazing in the last year.

That’s not to say that architectures aren’t interesting or important or that the improvements aren’t useful, but it is a little bit of a surprise, even though it shouldn’t be at this point because it’s probably just a version of the Bitter Lesson.

imjonse1 hour ago
> On the LLM Architecture Gallery, it’s interesting to see the variations between models, but I think the 30,000ft view of this is that in the last seven years since GPT-2 there have been a lot of improvements to LLM architecture but no fundamental innovations in that area.

After years of showing up in papers and toy models, hybrid architectures like Qwen3.5 contain one such fundamental innovation - linear attention variants which replace the core of transformer, the self-attention mechanism. In Qwen3.5 in particular only one of every four layers is a self-attention layer.

MoEs are another fundamental innovation - also from a Google paper.

libraryofbabel42 minutes ago
Thanks for the note about Qwen3.5. I should keep up with this more. If only it were more relevant to my day to day work with LLMs!

I did consider MoEs but decided (pretty arbitrarily) that I wasn’t going to count them as a truly fundamental change. But I agree, they’re pretty important. There’s also RoPE too, perhaps slightly less of a big deal but still a big difference from the earlier models. And of course lots of brilliant inference tricks like speculative decoding that have helped make big models more usable.

iroddis9 hours ago
This is amazing, such a nice presentation. It reminds me of the Neural Network Zoo [1], which was also a nice visualization of different architectures.

[1] https://www.asimovinstitute.org/neural-network-zoo/

bicepjai40 minutes ago
Currently working on a similar project for myself. This looks like a great resource. Thanks for sharing. https://llm-lab.bicepjai.com/
wood_spirit10 hours ago
Lovely!

Is there a sort order? Would be so nice to understand the threads of evolutions and revolution in the progression. A bit of a family tree and influence layout? It would also be nice to have a scaled view so you can sense the difference in sizes over time.

krackers9 hours ago
There is https://magazine.sebastianraschka.com/p/technical-deepseek which shows an evolution in deepseek family
andai1 hour ago
> The goal of the proof verifier (LLM 2) is to check the generated proofs (LLM 1), but who checks the proof verifier? To make the proof verifier more robust and prevent it from hallucinating issues, they developed a third LLM, a meta-verifier.
krackers36 minutes ago
The one thing I didn't quite understand (and wasn't mentioned in their paper unless I missed it), is why you can't keep stacking turtles. You probably get diminishing returns at some point, but why not have a meta-meta-verifier?
gasi8 hours ago
So cool — thanks for sharing! Here’s a zoomable version of the diagram: https://zoomhub.net/LKrpB
Slugcat7 hours ago
What tool was used to draw the diagrams?
nxobject6 hours ago
Thank you so much! As a (bio)statistician, I've always wanted a "modular" way to go from "neural networks approximate functions" to a high-level understanding about how machine learning practitioners have engineered real-life models.
LuxBennu7 hours ago
Interesting collection. The architecture differences show up in surprising ways when you actually look at prompt patterns across models. Longer context windows don't just let you write more, they change what kind of input structure works best.
jasonjmcghee5 hours ago
What's the structurally simplest architecture that has worked to a reasonably competitive degree?
loveparade5 hours ago
Competitiveness doesn't really come from architecture, but from scale, data, and fine-tuning data. There has been little innovation in architecture over the last few years, and most innovations are for the purpose of making it more efficient to run training or inference (fit in more data), not "fundamentally smarter"
bigyabai5 hours ago
If your definition of "competitive" is loose enough, you can write your own Markov chain in an evening. Transformer models rely on a lot of prior art that has to be learned incrementally.
jasonjmcghee5 hours ago
Not that loose lol.

I’m thinking it’s still llama / dense decoder only transformer.

travisgriggs6 hours ago
Darn. I clicked here hoping we were having LLMs design skyscrapers, dams, and bridges.

I even brought my popcorn :(

jrvarela566 hours ago
Would be awesome to see something like this for agents/harnesses
charcircuit7 hours ago
I'm surprised at how similar all of them are with the main differences being the size of layers.
arikrahman5 hours ago
Thank you for the high quality diagrams!
mvrckhckr9 hours ago
What a great idea and nice execution.
neuroelectron6 hours ago
An older post from this blog, the linked article was updated recently: https://news.ycombinator.com/item?id=44622608
jawarner2 hours ago
Looks like this may have received the HN Hug of Death. I'm getting "Too Many Requests" error trying to load the images.
brianjking2 hours ago
I'm getting that trying to load the content at all, text included.
FailMore11 hours ago
Thanks! This is cool. Can you tell me if you learnt anything interesting/surprising when pulling this together? As in did it teach you something about LLM Architecture that you didn't know before you began?