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

581–590 of 609 posts

Re: Gemini 3 Flash: Frontier intelligence built for speed

#581

Post by mark_l_watson »

thecupisblue wrote: post #97
fariszr wrote: post #7

These flash models keep getting more expensive with every release. Is there an OSS model that's better than 2.0 flash with similar pricing, speed and a 1m context window? Edit: this is not the typical flash model, it's actually an insane value if the benchmarks match real world usage. > Gemini 3 Flash achieves a score of 78%, outperforming not only the 2.5 series, but also Gemini 3 Pro. It strikes an ideal balance fo…

Yes, but the 3.0 Flash is cheaper, faster and better than 2.5 Pro. So if 2.5 Pro was good for your usecase, you just got a better model for about 1/3rd of the price, but might hurt the wallet a bit more if you use 2.5 Flash currently and want an upgrade - which is fair tbh.

I agree, adding one point: a better model can in effect use fewer tokens if you get a higher percentage of successful one-shots to work. I am a ‘retired gentleman scientist’ so take this with a grain of salt (I do a lot of non-commercial, non-production experiments): when I watch the output for tool use, better models have fewer tool ‘re-tries.’

Re: Gemini 3 Flash: Frontier intelligence built for speed

#582

Post by mark_l_watson »

sosodev wrote: post #286
fariszr wrote: post #7

These flash models keep getting more expensive with every release. Is there an OSS model that's better than 2.0 flash with similar pricing, speed and a 1m context window? Edit: this is not the typical flash model, it's actually an insane value if the benchmarks match real world usage. > Gemini 3 Flash achieves a score of 78%, outperforming not only the 2.5 series, but also Gemini 3 Pro. It strikes an ideal balance fo…

Nvidia released Nemotron 3 nano recently and I think it fits your requirements for an OSS model: https://huggingface.co/nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B... It's extremely fast on good hardware, quite smart, and can support up to 1m context with reasonable accuracy

I second this: I have spent about five hours this week experimenting with Nemotron 3 nano for both tool use and code analysis: it is excellent! and fast!

Relevant to the linked Google blog: I feel like getting Nemotron 3 nano and Gemini 3 flash in one week is an early Christmas gift. I have lived with the exponential improvements in practical LLM tools over the last three years, but this week seems special.

Re: Gemini 3 Flash: Frontier intelligence built for speed

#584

Post by mark_l_watson »

Davidzheng wrote: post #416
JumpCrisscross wrote: post #161

Earlier quoted context omitted.

> think about what would be most terrifying to Anthropic and OpenAI The most terrifying thing would be Google expanding its free tiers.

on aistudio the free tier limits on all models are decent

I turned on API billing on API Studio in the hope of getting the best possible service. As long as you are not using the Gemini thinking and research APIs for long-running computations, the APIs are very inexpensive to use.

Re: Gemini 3 Flash: Frontier intelligence built for speed

#585

Post by seunosewa »

mips_avatar wrote: post #288
samyok wrote: post #4

Don’t let the “flash” name fool you, this is an amazing model. I have been playing with it for the past few weeks, it’s genuinely my new favorite; it’s so fast and it has such a vast world knowledge that it’s more performant than Claude Opus 4.5 or GPT 5.2 extra high, for a fraction (basically order of magnitude less!!) of the inference time and price

OpenAI made a huge mistake neglecting fast inferencing models. Their strategy was gpt 5 for everything, which hasn't worked out at all. I'm really not sure what model OpenAI wants me to use for my applications that require lower latency. If I follow their advice in their API docs about which models I should use for faster responses I get told either use GPT 5 low thinking, or replace gpt 5 with gpt 4.1, or switch to…

GPT 5 Mini is supposed to be equivalent to Gemini Flash.

Re: Gemini 3 Flash: Frontier intelligence built for speed

#586

Post by thecupisblue »

lancekey wrote: post #387
thecupisblue wrote: post #78

Earlier quoted context omitted.

Oh wow - I recently tried 3 Pro preview and it was too slow for me. After reading your comment I ran my product benchmark against 2.5 flash, 2.5 pro and 3.0 flash. The results are better AND the response times have stayed the same. What an insane gain - especially considering the price compared to 2.5 Pro. I'm about to get much better results for 1/3rd of the price. Not sure what magic Google did here, but would love…

Curious to learn what a “product benchmark” looks like. Is it evals you use to test prompts/models? A third party tool? Examples from the wild are a great learning tool, anything you’re able to share is appreciated.

It's an internal benchmark that I use to test prompts, models and prompt-tunes, nothing but a dashboard calling our internal endpoints and showing the data, basically going through the prod flow.

For my product, I run a video through a multimodal LLM with multiple steps, combine data and spit out the outputs + score for the video.

I have a dataset of videos that I manually marked for my usecase, so when a new model drops, I run it + the last few best benchmarked models through the process, and check multiple things:

- Diff between outputed score and the manual one - Processing time for each step - Input/Output tokens - Request time for each step - Price of request

And the classic stats of average score delta, average time, p50, p90 etc. + One fun thing which is finding the edge cases, since even if the average score delta is low (means its spot-on), there are usually some videos where the abs delta is higher, so these usually indicate niche edge cases the model might have.

Gemini 3 Flash nails it sometimes even better than the Pro version, with nearly the same times as 2.5 Pro does on that usecase. Actually, pushed it to prod yesterday and looking at the data, it seems it's 5 seconds faster than Pro on average, with my cost-per-user going down from 20 cents to 12 cents.

IMO it's pretty rudimentary, so let me know if there's anything else I can explain.

Re: Gemini 3 Flash: Frontier intelligence built for speed

#587

Post by thecupisblue »

m00dy wrote: post #475
thecupisblue wrote: post #78

Earlier quoted context omitted.

Oh wow - I recently tried 3 Pro preview and it was too slow for me. After reading your comment I ran my product benchmark against 2.5 flash, 2.5 pro and 3.0 flash. The results are better AND the response times have stayed the same. What an insane gain - especially considering the price compared to 2.5 Pro. I'm about to get much better results for 1/3rd of the price. Not sure what magic Google did here, but would love…

May I ask your internal benchmark ? I'm building a new set of benchmarks and testing suite for agentic workflows using deepwalker [0]. How do you design your benchmark suite ? would be really cool if you can give more details. [0] https://deepwalker.xyz

Shared a bit more here - https://news.ycombinator.com/item?id=46314047.

But pretty rudimentary, nothing special. Also did not know about deepwalker, looks quite interesting - you building it?

Re: Gemini 3 Flash: Frontier intelligence built for speed

#588

Post by xzjis »

waffletower wrote: post #283
Workaccount2 wrote: post #262

Earlier quoted context omitted.

Open source models are riding coat tails, they are basically just distilling the giant SOTA models, hence perpetually being 4-6mos behind.

If this quantification of lag is anywhere near accurate (it may be larger and/or more complex to describe), soon open source models will be "simply good enough". Perhaps companies like Apple could be 2nd round AI growth companies -- where they market optimized private AI devices via already capable Macbooks or rumored appliances. While not obviating cloud AI, they could cheaply provide capable models without subscrip…

Google already has dedicated hardware for running private LLMs: just look at what they're doing on the Google Pixel. The main limiting factor right now is access to hardware that's powerful enough, and especially has enough memory, to run a good LLM, which will happen eventually. Normally, by 2031 we should have devices with 400 GB of RAM, but the current RAM crisis could throw off my calculations...

Re: Gemini 3 Flash: Frontier intelligence built for speed

#590

Post by metadat »

vanviegen wrote: post #530
mips_avatar wrote: post #156

Earlier quoted context omitted.

I'd be curious how many people use openrouter byok just to avoid figuring out the cloud consoles for gcp/azure.

Openrouter is great! Prepaid, no surprise bills. Easily switch between any models you desire. Dead simple interface. Reliable. What's not to like?

With OpenRouter it can be unclear if you're getting a quantized model or not.
Post reply on HN