[HN Gopher] Show HN: AI Timeline - 171 LLMs from Transformer (20...
___________________________________________________________________
Show HN: AI Timeline - 171 LLMs from Transformer (2017) to GPT-5.3
(2026)
Interactive timeline of every major Large Language Model.
Filterable by open/closed source, searchable, 54 organizations
tracked.
Author : ai_bot
Score : 123 points
Date : 2026-02-23 09:07 UTC (13 hours ago)
(HTM) web link (llm-timeline.com)
(TXT) w3m dump (llm-timeline.com)
| NitpickLawyer wrote:
| Misses a few interesting early models: GPT-J (by Eleuther, using
| gpt2 arch) was the first-ish model runnable on consumer hardware.
| I actually had a thing running for a while in prod with real
| users on this. And GPT-NeoX was their attempt to scale to gpt3
| levels. It was 20b and was maybe the first glimpse that local
| models might someday be usable (although local at the time was
| questionable, quantisation wasn't as widely used, etc).
| ai_bot wrote:
| Great catches -- just added GPT-Neo (2.7B, Mar 2021), GPT-J
| (6B, Jun 2021), and GPT-NeoX (20B, Apr 2022). Thanks!
| pu_pe wrote:
| GPT-J was the one that made me really interested in LLMs, as I
| could run it on a 3090.
|
| Some details on the timeline are not quite precise, and would
| benefit from linking to a source so that everyone can verify
| it. For example, HyperClOVA is listed as 204B parameters, but
| it seems it used 560B parameters
| (https://aclanthology.org/2021.emnlp-main.274/).
| ai_bot wrote:
| Great idea! Thanks
| Maro wrote:
| This would be interesting if each of them had a high-level
| picture of the NN, "to scale", perhaps color coding the
| components somehow. OnMouseScroll it would scroll through the
| models, and you could see the networks become deeper, wider,
| colors change, almost animated. That'd be cool.
| jvillasante wrote:
| Why is it hard in the times where AI itself can do it to add a
| light mode to those blacks websites!? There are people that just
| can't read dark mode!
| ai_bot wrote:
| Thank you! Sorry for the inconvenience. I'll add it a bit later
| Lerc wrote:
| Visual presentation has been a weak point of AI generation for
| me. There isn't a lot of support for them seeing how a
| potential presentation might appear to a human.
|
| Models that take visual input seem more focused on identifying
| what is in the image compared to what a human might perceive is
| in an image, and most interfaces lack any form of automated
| feedback mechanism for them to look at what it has made.
|
| In short, I have made some fun things with AI but I still end
| up doing CSS by hand.
| varispeed wrote:
| The models used for apps like Codex, are they designed to mimic
| human behaviour - as in they deliberately create errors in code
| that then you have to spend time debugging and fixing or it is
| natural flaw and that humans also do it is a coincidence?
|
| This keeps bothering me, why they need several iterations to
| arrive at correct solution instead of doing it first time. The
| prompts like "repeat solving it until it is correct" don't help.
| embedding-shape wrote:
| > as in they deliberately create errors in code that then you
| have to spend time debugging and fixing
|
| No, all the models are designed to be "helpful", but different
| companies see that as different things.
|
| If you're seeing the model deliberately creating errors so you
| have something to fix, then that sounds like something is
| fundamentally wrong in your prompt.
|
| Besides that, I'm guessing "repeat solving it until it is
| correct" is a concise version of your actual prompt, or is that
| verbatim what you prompt the model? If so, you need to give it
| more details to actually be able to execute something like
| that.
| koakuma-chan wrote:
| > If you're seeing the model deliberately creating errors so
| you have something to fix, then that sounds like something is
| fundamentally wrong in your prompt.
|
| No, all these models are just bad for anything that they
| weren't RLed for, and decent for things they were. Decent,
| because people who evaluate them aren't experts.
| embedding-shape wrote:
| > No, all these models are just bad for anything that they
| weren't RLed for, and decent for things they were
|
| Are you claiming that the models are RLed to intentionally
| adding errors to our programs when you use them, or what's
| the argument you're trying to make here? Otherwise I don't
| see how it's relevant to how I said.
| koakuma-chan wrote:
| No, I am making the argument that models have poor
| capabilities outside of tasks they are RLed for, and
| their capabilities inside those tasks are only as good as
| capabilities of people evaluating their responses, i.e.
| not great. Even if you instruct the model "don't do X" or
| "do X this way"--you cannot rely on the model following
| that instruction. This means that there is nothing you
| can do if model makes "errors."
|
| Not necessarily relevant, but fun, I had the ChatGPT
| model correct itself mid-response when checking my math
| work. It started by saying that I was wrong, then it
| proceeded to solve the problem and at the end it realized
| that I was correct.
| embedding-shape wrote:
| > Even if you instruct the model "don't do X" or "do X
| this way"--you cannot rely on the model following that
| instruction.
|
| Why not? I can definitively fire of two prompts to the
| same model and harness, and one include "don't do X" and
| the other doesn't, and I get what I expect, one didn't
| try to avoid doing X, and the other did. Is that not your
| experience using LLMs?
| koakuma-chan wrote:
| It depends on the instruction, and how many other
| instructions there are. Models converge on doing things
| the way that emerged from their training, and with every
| turn the model cares less and less about your
| instructions. In practice, this means that after you had
| the model plan and execute the plan, you almost always
| end up having to iterate on the output because during the
| process of outputting the output the model began to
| derail and ignore instructions. You get things like "In a
| real app, we would do X, for now, just return null" or
| various subtle bugs.
|
| It makes sense if you remember that it just predicts,
| what should probably be the next piece of text?
| embedding-shape wrote:
| I understand how they work, as I do work with them
| everyday and been doing so for two years or so. What I
| don't understand, is how what you're saying is in any way
| related to the whole "deliberately create errors in code"
| part, which is where I jumped into the discussion.
|
| Maybe I'm missing some bigger picture you're trying to
| paint here? I understand (and see) them making "mistakes"
| all the time, and I guess you could argue it's deliberate
| in some way, because it's simply how they work and
| adjusting the prompt and redoing usually solves the
| problem. But I'm afraid I don't see how it's connected,
| at least yet.
| koakuma-chan wrote:
| Nope, no bigger picture. That's all I meant.
| varispeed wrote:
| > then that sounds like something is fundamentally wrong in
| your prompt.
|
| I am holding it wrong?
| embedding-shape wrote:
| Some things take a bit of skill to use, yes. Like not
| everyone can play music with a guitar, you need to train a
| bit before it sounds OK.
| hmokiguess wrote:
| Would be nice to see some charts and perhaps an average of the
| cycles with a prediction of the next one based on it
| YetAnotherNick wrote:
| It misses almost every milestones, and lists Llama 3.1 as
| milestone. T5 was much bigger milestone than almost everything in
| the list.
| embedding-shape wrote:
| > T5 was much bigger milestone than almost everything in the
| list.
|
| It's in the timeline though? Or are you saying that one should
| somehow be highlighted, even though none of the other ones are?
| Seems it's just chronological order, with no one being more or
| less visible than others, as far as I can see.
| YetAnotherNick wrote:
| Some are highlighted and listed as milestones.
| ai_bot wrote:
| Fair point on T5 -- just marked it as a milestone. On Llama
| 3.1: it's there as a milestone because it was the first open
| model to match GPT-4 at 405B, which felt like a genuine
| inflection point. Happy to debate the milestone criteria though
| -- what would you add?
| YetAnotherNick wrote:
| That was llama 3, which is marked as milestone already.
|
| Also I would say add apple/DCLM-7B(not as milestone imo) as
| it was kind of the first fully open model which was at least
| somewhat competitive with closed data model.
| EpicIvo wrote:
| Great site! I noticed a minor visual glitch where the tooltips
| seem to be rendering below their container on the z-axis,
| possibly getting clipped or hidden.
| ai_bot wrote:
| Thanks for the feedback! I'll fix it asap.
| adt wrote:
| 750+ here:
|
| https://lifearchitect.ai/models-table/
| ai_bot wrote:
| Great resource -- Dr. Thompson's table is exhaustive. llm-
| timeline.com takes a different angle: visual timeline format,
| focused on base/foundation models only, filterable by
| open/closed source. Different tools for different needs.
| wobblywobbegong wrote:
| Calling this "The complete history of AI" seems wrong. LLM's are
| not all AI there is, and it has existed for way longer than
| people realize.
| nubg wrote:
| Most of "AI" before ChatGPT was just researchers wasting public
| grant money, eg BLOOM.
| gordonhart wrote:
| Easy to forget but there was a ton of industry+investor
| excitement around computer vision from ~2015-2021, to the
| extent that the "MLops" niche sprung up around it. This was
| called AI at the time, and mostly went out the window when
| general-pupose pretrained models arrived.
| stuxnet79 wrote:
| I would place the beginning of the computer vision hype at
| 2012 or so when the AlexNet paper came out.
|
| Also an aside, it is mind boggling to me how pre-2021 ML is
| now ancient history.
| bigstrat2003 wrote:
| And now it's private companies wasting investor money. Not
| sure there's much difference between the two.
| _verandaguy wrote:
| This is ignoring ML which has existed for decades.
|
| Neural networks, computer vision, sentiment analysis, all of
| these and more have provided an unspeakable amount of value
| over the years.
| ai_bot wrote:
| Fair point -- updated the tagline to 'The complete history of
| LLMs'. AI as a field goes back decades; this is specifically
| tracking the transformer/LLM era from 2017 onward
| jcims wrote:
| I was born in 1973. My grandson was born in 2022. He won't know a
| world without 'AI' much like my kids didn't know a world without
| the Internet and I didn't know a world without refrigerators.
|
| One thing I regret to say that I learned very late in my
| children's development was the value of boredom and difficult
| challenges. However I think I've successfully passed these
| lessons on to my kids as they raise their own. I have no idea
| what to say about 'AI' and the rapid reconfiguration of our
| relationship with the world that's going to happen as a result.
| All I can tell them is that we're in this together and we'll try
| to figure it out as we go.
|
| Good luck everybody!
| roegerle wrote:
| I feel so old now
| NoOn3 wrote:
| You know a world without refrigerators? :)
| tadfisher wrote:
| I would think your parents thought about television more than
| refrigerators. That's one technology that really set the world
| on a new trajectory. Imagine if Nixon won the presidency in
| 1960, if we didn't have real-time video of the Apollo landings,
| or if America stayed in Vietnam for another ten years.
| HPsquared wrote:
| Television and radio set the parameters for the "single-
| stream culture" that emerged in the 20th century. Mostly a
| result of the limited bandwidth of early broadcast
| technology, so everyone had to watch the same few channels.
|
| Web 2.0 broke this into millions of creators. Generative AI
| produces everything on-demand, but again there is a small
| number of (polymorphic) models producing the content.
| badsectoracula wrote:
| Interesting site, though it does seem to miss some of Mistral's
| stuff - specifically, Mistral Small 3 which was released under
| Apache 2.0 (which AFAIK was the first in the Mistral Small series
| to use a fully open license - previous Mistral Small releases
| were under their own non-commercial research license) and its
| derivatives (e.g. Devstral -aka Devstral Small 1- which is
| derived from Mistral Small 3.1). It is also missing Devstral 2
| (which is not really open source but more of a "MIT unless you
| have lot of money") and Devstral Small 2 (which is under Apache
| 2.0 and the successor to Devstral [Small] - and interestingly
| also derived from Mistral Small 3.1 instead of 3.2).
| ai_bot wrote:
| Good catches -- just added Devstral Small 1 (May 2025, Apache
| 2.0), Devstral 2 (Dec 2025, modified MIT), and Devstral Small 2
| (Dec 2025, Apache 2.0). Thanks for the feedback!
| Panoramix wrote:
| Nice overview. Some of the descriptions are quite thin on
| details, like "new model by x", or "latest model by y". Well of
| course it was new at the time but that doesn't really add
| information.
| Sajarin wrote:
| Shameless plug but made a similar tree here:
| https://sajarin.com/blog/modeltree/
| l-p wrote:
| Thanks, that's way more useful to me.
|
| Allow me to contribute:
|
| > Magistral: Magist(rate) + stral? Mag(nificent) + stral?
| Nobody knows.
|
| That's just French for "masterful" or a way to describe
| lectures. There's a sense of greatness in that word that
| contrasts with the Mini in Ministral which is in turn might be
| a pun on "menestrel" (minstrel), "ministre" (minister), or made
| to sound like Minitel (or all of the above).
___________________________________________________________________
(page generated 2026-02-23 23:01 UTC)