News for hackers. Stuff that matters.
Re:Gemini 4 Neon by onlyrealcuzzo on Wednesday November 18, 2026 @04:46PM (#50552601)
I suspect they may want to get away from the size class due to the fact that - at the pace they are launching new models - and the similarity of model sizes between Neon/Carbon, Opus/Fable, Sol/Astra - the best model is going to be the most recently released one, not the biggest one.

I'm not sure it's much better than 4.1 Flash vs 4 Pro -> if 4.1 Flash is better temporarily when Pro is typically significantly better.

I'm not really sure how you get around this problem easily.

Re:Gemini 4 Neon by nicoburns on Wednesday November 18, 2026 @04:49PM (#50552651)
Interesting. You're claiming that the Opus/Sol/Neon class and Fable/Astra/Carbon class models are similar sizes to each other? That would be surprising to me.
Re:Gemini 4 Neon by dannyw on Wednesday November 18, 2026 @04:56PM (#50552756)
A Microsoft Azure page briefly claimed Sol runs with less looped transformer passes; and so you could speculate that Sol and Astra might share weights; but have different “loop” factors.

That would make a lot of sense from an inference optimisation perspective. The same prefill and decode machines can hold one set of weights, and serve both Astra and Sol requests.

With MoEs, there’s probably stuff you can play around with varying the number of active experts.

Re:Gemini 4 Neon by nicoburns on Wednesday November 18, 2026 @04:59PM (#50552792)
I heard that Sol 6.1 was cut down Astra because Sol 6.0 was received badly. But I thought that was the exception (and a result of Open AI's actual Sol-class model being uncompetitive) rather than the rule.
Re:Gemini 4 Neon by willy_k on Wednesday November 18, 2026 @05:15PM (#50553019)
That would line up - on release and for the few following weeks Sol 6.1 felt like Astra but maybe more task oriented, recently it’s been a noticeable downgrade in the level of specification needed to successfully set off.
Re:Gemini 4 Neon by onlyrealcuzzo on Wednesday November 18, 2026 @05:09PM (#50552937)
> A Microsoft Azure page briefly claimed Sol runs with less looped transformer passes;

This doesn't really make sense given how the models released.

OpenAI released Astra long before latent-space looping was a thing at the frontier space.

Then they released Sol 6, which was fast, clearly not looped. Then they release Sol 6.1 which was nearly as good as Astra, but way cheaper, and WAY slower -> this would follow it just being a looped version of Sol 6.

It doesn't seem likely that Astra 6.2 is just a looped Astra 6... It's too fast, and they'd go bankrupt if they're charging the same price... It may be a further looped version of Sol than Sol 6.2 - which would explain why it's not as much better as we would've hoped.

It may also explain why their release is not going as well as they hoped. It's hard to predict how your first looped models are going to fare.

Re:Gemini 4 Neon by iamdelirium on Wednesday November 18, 2026 @05:19PM (#50553070)
Looped transformers have been known for years now. I don't think it's something new.
Re:Gemini 4 Neon by onlyrealcuzzo on Wednesday November 18, 2026 @05:21PM (#50553107)
They are JUST arriving to frontier models.

Even in the open weight Chinese models as small as DeepSeek V4.1 Flash - nothing is looped.

Re:Gemini 4 Neon by refulgentis on Wednesday November 18, 2026 @05:28PM (#50553207)
I think they're responding to the claim Astra can't be looped, there's clear and convincing research, as well as public comms, that it is. It's unclear why Astra can't be looped because looping is new, but, 6.1 Sol released a week later can be.
Re:Gemini 4 Neon by refulgentis on Wednesday November 18, 2026 @05:26PM (#50553194)
Astra was the first shipping looper. Unconvincing evidence would be reporting x OpenAI's quick reaction to it via a blog post from their chief research (it broke what's considered a safety taboo). Convincing evidence would be research from ex. Neel Nanda on serial reasoning, recent update at https://x.com/EpochAIResearch/status/2122762411287748696?a=J...
Re:Gemini 4 Neon by onlyrealcuzzo on Wednesday November 18, 2026 @05:03PM (#50552847)
Fable & Opus, Astra & Sol, and Carbon are all thought to be >1T param models - AFAIK, most people think Opus & Sol are ~5T param models and Astra, Fable are 10T+.

I think the rumor-mill consensus is that Google's models are significantly smaller while attempting to swing at the same level - but the ratios AFAIK are thought to be around the same.

Re:Gemini 4 Neon by ml-anon on Wednesday November 18, 2026 @05:24PM (#50553148)
So many figures and ratios pulled from a single ass
Re:Gemini 4 Neon by andai on Wednesday November 18, 2026 @05:07PM (#50552900)
The Bitter Lesson - Richard Sutton (2019)

https://news.ycombinator.com/item?id=19393432

Re:Gemini 4 Neon by weitendorf on Wednesday November 18, 2026 @05:32PM (#50553263)
There's only a limited amount of hardware for training and serving models at scale, and which form an upper bound for what model size/inference can support.

A GB300 rack has 72x288GB=20TB VRAM. Do the math and 10T parameter models are just about the max size you can easily train and serve (at 1M context) without doing a lot of custom hardware development or limiting your scale.