Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Gemini 3.5 Flash

blog.google

Post reply on HN
Print view View original

571–580 of 692 posts

Re: Gemini 3.5 Flash

#571

Post by 4ggr0 »

easygenes wrote: post #466

For those who would like to know the total and active parameter count of this model: even though Google doesn't disclose the model technicals, we can infer them within relatively tight margins based on what we do know. We know they serve the model on TPU 8i, which we have plenty of hard specs for (so we know the key constraints: total memory and bandwidth and compute flops). We can also set a ceiling on the compute c…

meta - i think that's the first time i've seen a table in a hn comment, and i'm surprised/impressed! nice

are these pre-generated in a different tool with plain unicode and then just copy-pasted, or is it a built-in feature of hn?

Re: Gemini 3.5 Flash

#572

Post by pimeys »

No computer use yet. I wonder when they enable it for this model, CUA was one of the main selling points for us with the previous version of Flash.

Re: Gemini 3.5 Flash

#573

Post by znpy »

CobrastanJorji wrote: post #314
nightski wrote: post #189

Earlier quoted context omitted.

Very little of what made the Flash culture so fun made its way into HTML5.

I dunno, the tools are kind of there. Browsers have canvases and JavaScript and SVGs and sound. The communities are around; they're just kind of dispersed. There's no one website that is THE place for fun stuff. Instead, there are dozens, and most of them suck. There's still fun stuff, though. I stumbled upon this bit of insanity just yesterday: https://tykenn.itch.io/trees-hate-you . It would have fit in fabulously…

The issue is that flash had everything you mention about fifteen to twenty years ago (if more) along with better and more thorough tooling.

In the html5 camp the features appeared one by one and the tooling is still fragmented.

What happened between flash dying and html5 having a complete toolset is that interest died.

Re: Gemini 3.5 Flash

#574

Post by mrbungie »

kristopolous wrote: post #438
mrbungie wrote: post #433

Earlier quoted context omitted.

It is kind of noisy because the release recency, which is what your "age" column actually represents, is not important data for the comparison you are trying to make. Also what message we should get from that table is not really obvious.

Okay I think there's a familiarity delta. I constantly run into this I know artificial analysis quite well as the gold standard in llm evals. But I guess they're still obscure I didn't think they were. The age is important because new techniques keep being developed and so it is a very rough indicator of the size/cost/efficiency trade-off. How old a model is is a major indicator of what you can expect from it. I real…

> I know artificial analysis quite well as the gold standard in llm evals.

I also know them, but it took me a while to realise you were publishing their data in that table. I don't think it was clear.

> The age is important because new techniques keep being developed and so it is a very rough indicator of the size/cost/efficiency trade-off.

Yes but you are already including the name of the model, your potential public for the table already know about model's release history and therefore each model's age, at least roughly.

Re: Gemini 3.5 Flash

#575

Post by ashirviskas »

GodelNumbering wrote: post #84

Per million input/output tokens: Gemini 2.5 flash: $0.30/$2.50 Gemini 3.0 flash preview: $0.50/$3.00 Gemini 3.5 flash: $1.50/$9.00 Interesting pricing direction. I don't think we have ever seen a 3x price increase for in the immediate next same-sized model (and lol @ 3 only ever getting a preview). 3.5 flash costs similar to Gemini 2.5 pro which was $1.25/$10

don't forget Gemini 2.0 flash at $0.10/$0.40

Re: Gemini 3.5 Flash

#576

Post by ashirviskas »

__jl__ wrote: post #263
GodelNumbering wrote: post #84

Per million input/output tokens: Gemini 2.5 flash: $0.30/$2.50 Gemini 3.0 flash preview: $0.50/$3.00 Gemini 3.5 flash: $1.50/$9.00 Interesting pricing direction. I don't think we have ever seen a 3x price increase for in the immediate next same-sized model (and lol @ 3 only ever getting a preview). 3.5 flash costs similar to Gemini 2.5 pro which was $1.25/$10

This understates the cost increase. 3.5 Flash also uses more tokens. artificialanalysis.ai shows these difference to run the whole eval, which I think is more realistic pricing: Gemini 2.5 flash (27 score): $172 (1.0x) Gemini 2.5 pro (35 score): $649 (3.8x) Gemini 3.0 Flash (46 score): $278 (1.6x) Gemini 3.5 Flash (55 score): $1,552 (9.0x or 2.4x compared to 2.5 pro) This is a massive price increase... 5.6x compared…

Gemini 2.0 Flash: $19

Re: Gemini 3.5 Flash

#577

Post by Tade0 »

lmazgon wrote: post #567

Click on "Listen to article", make sure the voice is "Umbriel" and skip to 4:15 - there's a hallucinated part at the end in Russian (I think). On a blog post about the latest and greatest AI model. Oh the irony.

I ran it through speech-to-text and it starts with something among the lines of "dear colleagues, just like a doctor tells a patient 'health can wait'...", after that it's nonsense.

I don't know if what the doctor said is some kind of idiomatic expression, but appears to be the opposite of sound medical advice. :)

Re: Gemini 3.5 Flash

#578

Post by nomilk »

simonw wrote: post #131

The pelican is a lot : https://github.com/simonw/llm-gemini/issues/133#issuecomment... Not a great bicycle though, it forgot the bar between the pedals and the back wheel and weirdly tangled the other bars. Expensive too - that pelican cost 13 cents: https://www.llm-prices.com/#it=11&ot=14403&sel=gemini-3.5-fl...

'Pelicans' should be the unit of measurement for model prices, rather than tokens.

Re: Gemini 3.5 Flash

#579

Post by kzrdude »

holtkam2 wrote: post #174
simonw wrote: post #131

The pelican is a lot : https://github.com/simonw/llm-gemini/issues/133#issuecomment... Not a great bicycle though, it forgot the bar between the pedals and the back wheel and weirdly tangled the other bars. Expensive too - that pelican cost 13 cents: https://www.llm-prices.com/#it=11&ot=14403&sel=gemini-3.5-fl...

at a certain point you're gonna need to change your benchmark because this will end up in the model's training set

As mentioned in another recent thread, that time is now.

Re: Gemini 3.5 Flash

#580

Post by cloakandswagger »

LetsGetTechnicl wrote: post #258
roadside_picnic wrote: post #199

Earlier quoted context omitted.

These companies are unprofitable (as all companies at this stage and ambition should be) but I increasingly don't see any justification for the idea that it is fundamentally unprofitable. Inference alone is certainly profitable. I'm running models at home that are comparable to performance of paid models a year or so ago for free. Even for much larger models the cost around inference serving are clearly manageable. T…

If it's profitable, why haven't they reported any profits? People like Ed Zitron have done the math and it just doesn't add up. I mean he just published this piece today: https://www.wheresyoured.at/ai-is-too-expensive/

Zitron thinks capex is a liability that needs to be paid off in a year instead of a long-standing asset.

Similarly he thinks that an investment into an AI startup is also a loan that the startup needs to pay back out of their own revenue, instead of a share of a company that will IPO at a higher valuation.

Basically his doomerism is a byproduct of financial illiteracy.

Post reply on HN