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

Gemini 3 Flash: Frontier intelligence built for speed

blog.google

Post reply on HN
Print view View original

551–560 of 609 posts

Re: Gemini 3 Flash: Frontier intelligence built for speed

#551

Post by hu3 »

floundy wrote: post #414
willis936 wrote: post #405

Earlier quoted context omitted.

That's too bad. Apple's most interesting value proposition is running local inference with big privacy promises. They wouldn't need to be the highest performer to offer something a lot of people might want.

Apple’s most interesting value proposition was ignoring all this AI junk and letting users click “not interested” on Apple Intelligence and never see it again. From a business perspective it’s a smart move (inasmuch as “integrating AI” is the default which I fundamentally disagree with) since Apple won’t be left holding the bag on a bunch of AI datacenters when/if the AI bubble pops. I don’t want to lose trust in App…

Sure. If by ignore you mean flaunt about Apple Intelligence only to fail miserably on the expectation they themselves generated.

Re: Gemini 3 Flash: Frontier intelligence built for speed

#552

Post by Simon321 »

qnleigh wrote: post #383

This model is breaking records on my benchmark of choice, which is 'the fraction of Hacker News comments that are positive.' Even people who avoid Google products on principle are impressed. Hardly anyone is arguing that ChatGPT is better in any respect (except brand recognition).

i don't know, chat gpt seems to hallucinate a lot less

Re: Gemini 3 Flash: Frontier intelligence built for speed

#553

Post by ygouzerh »

Palmik wrote: post #493
jdanbrown wrote: post #463

Earlier quoted context omitted.

Have you tried OpenRouter ( https://openrouter.ai )? I’ve been happy using it as a unified api provider with great model coverage (including Google, Anthropic, OpenAI, Grok, and the major open models). They charge 5% on top of each model’s api costs, but I think it’s worth it to have one centralized place to insert my money and monitor my usage. I like being able to switch out models without having to change my tools…

Why would you use OpenRouter rather than some local proxy like LiteLLM? I don't see the point of sharing data with more third parties and paying for the privilege. Not to mention that for coding, it's usually more cost efficient to get whatever subscription the specific model provider offers.

Thanks, I didn't knew about LiteLLM!

OpenRouter have some interesting providers, like Cerebras, which delivers 2,300 token/s on gpt-oss

Re: Gemini 3 Flash: Frontier intelligence built for speed

#555

Post by tomashubelbauer »

`gemini update` - error `gemini` and then `/update` - unknown command

I also had similar issues with Claude Code in the past. Everyone should take a page out of Bun's playbook. I never had `bun update` fail.

Edit: Also, I wish NPM wasn't the distribution mechanism for these TUIs. I suspect NPM's interplay with global packages and macOS permissions is what's causing the issue.

Re: Gemini 3 Flash: Frontier intelligence built for speed

#556

Post by Tiberium »

andai wrote: post #384
Tiberium wrote: post #243

Earlier quoted context omitted.

You can still set thinking budget to 0 to completely disable reasoning, or set thinking level to minimal or low.

>You cannot disable thinking for Gemini 3 Pro. Gemini 3 Flash also does not support full thinking-off, but the minimal setting means the model likely will not think (though it still potentially can). If you don't specify a thinking level, Gemini will use the Gemini 3 models' default dynamic thinking level, "high". https://ai.google.dev/gemini-api/docs/thinking#levels

I was talking about Gemini 3 Flash, and you absolutely can disable reasoning, just try sending thinking budget: 0. It's strange that they don't want to mention this, but it works.

Re: Gemini 3 Flash: Frontier intelligence built for speed

#557

Post by jack_riminton »

user34283 wrote: post #539
jack_riminton wrote: post #302

Earlier quoted context omitted.

But you’re forgetting the Jonny Ive hardware device that totally isn’t like that laughable pin badge thing from Humane /s

I agree completely. Altman was at some point talking about a screen less device and getting people away from the screen. Abandoning our mose useful sense, vision, is a recipe for a flop.

I'm not entirely sure it will ever see the light of day tbh

The amount of money sloshing around in these acquisitions makes you wonder what they're really for

Re: Gemini 3 Flash: Frontier intelligence built for speed

#558

Post by lobsterthief »

theshrike79 wrote: post #549
ggsp wrote: post #523

Earlier quoted context omitted.

Any suggestions for a simple tool to set up your own local evals?

My "tool" is just prompts saved in a text file that I feed to new models by hand. I haven't built a bespoke framework on top of it. ...yet. Crap, do I need to now? =)

_Brain backlog project #77 created_

Re: Gemini 3 Flash: Frontier intelligence built for speed

#559

Post by Europas »

katzenversteher wrote: post #487
lambda wrote: post #360

Earlier quoted context omitted.

The thing is, I see people use it for tricky niche knowledge all the time; using it as an alternative to doing a Google search. So I want to have a general idea of how good it is at this. I found something that was niche, but not super niche; I could easily find a good, human written answer in the top couple of results of a Google search. But until now, all LLM answers I've gotten for it have been complete hallucinat…

I also use niche questions a lot but mostly to check how much the models tend to hallucinate. E.g. I start asking about rank badges in Star Trek which they usually get right and then I ask about specific (non existing) rank badges shaped like strawberries or something like that. Or I ask about smaller German cities and what's famous about them. I know without the ability to search it's very unlikely the model actuall…

I'm waiting for properly adjusted specific LLMs. A LLM trained on so much trustworth generic data that it is able to understand/comprehend me and different lanugages but always talks to a fact database in the background.

I don't need an LLM to have a trillion parameters if i just need it to be a great user interface.

Someone is probably working on this somewere or will but lets see.

Post reply on HN