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

Gemini 3.1 Flash-Lite: Built for intelligence at scale

blog.google

Post reply on HN
Print view View original

21–30 of 34 posts

Re: Gemini 3.1 Flash-Lite: Built for intelligence at scale

#21

Post by vlmutolo »

Lots of comments about the price change, but Artifical Analysis reports that 3.1 Flash-Lite (reasoning) used fewer than half of the tokens of 2.5 Flash-Lite (reasoning).

This will likely bring the cost below 2.5 flash-lite for many tasks (depends on the ratio of input to output tokens).

That said, AA also reports that 3.1 FL was 20% more expensive to run for their complete Intelligence index benchmark.

The overall point is that cost is extremely task-dependent, and it doesn’t work to just measure token cost because reasoning can burn so many tokens, reasoning token usage varies by both task and model, and similarly the input/output ratios vary by task.

Re: Gemini 3.1 Flash-Lite: Built for intelligence at scale

#22

Post by xnx »

rohansood15 wrote: post #8

For the last 2 years, startup wisdom has been that models will continue to get cheaper and better. Claude first, and now Gemini has shown that it's not the case. We priced an enterprise contract using Flash 1.5 pricing last summer, and today that contract would be unit economic negative if we used Flash 3. Flash 2.5 and now Flash 3.1 Lite barely breaks even. I predict open-source models and fine-tuning are going to m…

> We priced an enterprise contract using Flash 1.5 pricing last summer,

Interesting. Flash 1.5 was already a year old at that point.

Re: Gemini 3.1 Flash-Lite: Built for intelligence at scale

#23

Post by simianwords »

rohansood15 wrote: post #16
simianwords wrote: post #13

Earlier quoted context omitted.

Not true. You just measure cost by amount of money spent per task. I would argue that this lite version is equivalent to older flash.

Yea but there is a whole world of tasks for which Flash 2.5-lite was sufficiently intelligent. Given Google's depreciation policy, there will soon be no way to get that intelligence at that price.

I hope they release models at every intelligence resolution although the thinking effort can be a good alternative

Re: Gemini 3.1 Flash-Lite: Built for intelligence at scale

#24

Post by k9294 »

stri8ted wrote: post #19
k9294 wrote: post #12

You can test Gemini 3.1 Lite transcription capabilities in https://ottex.ai — the only dictation app supporting Gemini models with native audio input. We benchmarked it for real-life voice-to-text use cases: Key takeaways: - 1.8x faster than Gemini 3 Flash on average - ~1.4 sec transcription time for short to medium recordings - ~$0.50/mo for heavy users (10h+ transcription) - Close to SOTA audio understanding and fo…

Can you show some comparisons for WER and other ASR models? Especially for non english.

I've been experimenting with Gemini 3.1 Flash Lite and the quality is very good.

I haven't found official benchmarks yet, but you can find Gemini 3 Flash word error rate benchmarks here: https://artificialanalysis.ai/speech-to-text/models/gemini — they are close to SOTA.

I speak daily in both English and Russian and have been using Gemini 3 Flash as my main transcription model for a few months. I haven't seen any model that provides better overall quality in terms of understanding, custom dictionary support, instruction following, and formatting. It's the best STT model in my experience. Gemini 3 Flash has somewhat uncomfortable latency though, and Flash Lite is much better in this regard.

Re: Gemini 3.1 Flash-Lite: Built for intelligence at scale

#25

Post by k9294 »

simianwords wrote: post #14
k9294 wrote: post #12

You can test Gemini 3.1 Lite transcription capabilities in https://ottex.ai — the only dictation app supporting Gemini models with native audio input. We benchmarked it for real-life voice-to-text use cases: Key takeaways: - 1.8x faster than Gemini 3 Flash on average - ~1.4 sec transcription time for short to medium recordings - ~$0.50/mo for heavy users (10h+ transcription) - Close to SOTA audio understanding and fo…

You know what would be great? A light weight wrapper model for voice that can use heavier ones in the background. That much is easy but what if you could also speak to and interrupt the main voice model and keep giving it instructions? Like speaking to customer support but instead of putting you on hold you can ask them several questions and get some live updates

It's actually a nice idea - an always-on micro AI agent with voice-to-text capabilities that listens and acts on your behalf.

Actually, I'm experimenting with this kind of stuff and trying to find a nice UX to make Ottex a voice command center - to trigger AI agents like Claude, open code to work on something, execute simple commands, etc.

Re: Gemini 3.1 Flash-Lite: Built for intelligence at scale

#26

Post by msp26 »

What the fuck is this price hike? It was such a nice low end, fast model. Who needs 10 years of reasoning on this model size??

I'm gonna switch some workflows to qwen3.5.

There's a lot of tasks that benefit from just having a mildly capable LLM and 2.5 Flash Lite worked out of the box for cheap.

Can we get flash lite lite please?

Edit: Logan said: "I think open source models like Gemma might be the answer here"

Implying that they're not interested in serving lower end Gemini models?

Re: Gemini 3.1 Flash-Lite: Built for intelligence at scale

#27

Post by msp26 »

vlmutolo wrote: post #21

Lots of comments about the price change, but Artifical Analysis reports that 3.1 Flash-Lite (reasoning) used fewer than half of the tokens of 2.5 Flash-Lite (reasoning). This will likely bring the cost below 2.5 flash-lite for many tasks (depends on the ratio of input to output tokens). That said, AA also reports that 3.1 FL was 20% more expensive to run for their complete Intelligence index benchmark. The overall po…

many tasks don't need any reasoning

Re: Gemini 3.1 Flash-Lite: Built for intelligence at scale

#28

Post by XCSme »

vlmutolo wrote: post #21

Lots of comments about the price change, but Artifical Analysis reports that 3.1 Flash-Lite (reasoning) used fewer than half of the tokens of 2.5 Flash-Lite (reasoning). This will likely bring the cost below 2.5 flash-lite for many tasks (depends on the ratio of input to output tokens). That said, AA also reports that 3.1 FL was 20% more expensive to run for their complete Intelligence index benchmark. The overall po…

> 3.1 Flash-Lite (reasoning)

(reasoning) doesn't say much. Is it low/med/high reasoning? I ran my own benchmarks, and 3.1 Flash-Lite on high costs A LOT: https://aibenchy.com/compare/google-gemini-3-1-flash-lite-pr...

Do not use 3.1 Flash-Lite with HIGH reasoning, it reasons for almost max output size, you can quickly get to millions of tokens of reasoning in a few requests.

Re: Gemini 3.1 Flash-Lite: Built for intelligence at scale

#29

Post by vlmutolo »

XCSme wrote: post #28
vlmutolo wrote: post #21

Lots of comments about the price change, but Artifical Analysis reports that 3.1 Flash-Lite (reasoning) used fewer than half of the tokens of 2.5 Flash-Lite (reasoning). This will likely bring the cost below 2.5 flash-lite for many tasks (depends on the ratio of input to output tokens). That said, AA also reports that 3.1 FL was 20% more expensive to run for their complete Intelligence index benchmark. The overall po…

> 3.1 Flash-Lite (reasoning) (reasoning) doesn't say much. Is it low/med/high reasoning? I ran my own benchmarks, and 3.1 Flash-Lite on high costs A LOT: https://aibenchy.com/compare/google-gemini-3-1-flash-lite-pr... Do not use 3.1 Flash-Lite with HIGH reasoning, it reasons for almost max output size, you can quickly get to millions of tokens of reasoning in a few requests.

Wow, that’s very interesting. I wish more benchmarks were reported along with the total cost of running that benchmark. Dollars per token is kind of useless for the reasons you mentioned.

Re: Gemini 3.1 Flash-Lite: Built for intelligence at scale

#30

Post by XCSme »

vlmutolo wrote: post #29
XCSme wrote: post #28

Earlier quoted context omitted.

> 3.1 Flash-Lite (reasoning) (reasoning) doesn't say much. Is it low/med/high reasoning? I ran my own benchmarks, and 3.1 Flash-Lite on high costs A LOT: https://aibenchy.com/compare/google-gemini-3-1-flash-lite-pr... Do not use 3.1 Flash-Lite with HIGH reasoning, it reasons for almost max output size, you can quickly get to millions of tokens of reasoning in a few requests.

Wow, that’s very interesting. I wish more benchmarks were reported along with the total cost of running that benchmark. Dollars per token is kind of useless for the reasons you mentioned.

Yup, MiniMax M-2.5 is a standout in that aspect. It's $/token is very low, because it reasons forever (fun fact, that's also the reason why it's #1 on OpenRouter, because it simply burns through tokens, and OpenRouter ranking is based on tokens usage)...
Post reply on HN