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

Improved Gemini 2.5 Flash and Flash-Lite

developers.googleblog.com

Post reply on HN
Print view View original

121–130 of 285 posts

Re: Improved Gemini 2.5 Flash and Flash-Lite

#121

Post by typpilol »

breakingcups wrote: post #84
jjani wrote: post #42

Earlier quoted context omitted.

Can't agree with that. Gemini doesn't lead just on price/performance - ironically it's the best "normie" model most of the time, despite it's lack of popularity with them until very recent. It's bad at agentic stuff, especially coding. Incomparably so compared to Claude and now GPT-5. But if it's just about asking it random stuff, and especially going on for very long in the same conversation - which non-tech users h…

My pet theory without any strong foundation is because OpenAI and Anthropic have trained their models really hard to fit the sycophantic mold of: =============================== Got it — *compliment on the info you've shared*, *informal summary of task*. *Another compliment*, but *downside of question*. ---------- (relevant emoji) Bla bla bla 1. Aspect 1 2. Aspect 2 ---------- *Actual answer* ----------- (checkmark e…

Anthropic also injects these long conversation reminders that are paragraph upon paragraphs about safety and what not to do.

People have said it destroys the intelligence mid convo

Re: Improved Gemini 2.5 Flash and Flash-Lite

#122

Post by typpilol »

porridgeraisin wrote: post #120
breakingcups wrote: post #84

Earlier quoted context omitted.

My pet theory without any strong foundation is because OpenAI and Anthropic have trained their models really hard to fit the sycophantic mold of: =============================== Got it — *compliment on the info you've shared*, *informal summary of task*. *Another compliment*, but *downside of question*. ---------- (relevant emoji) Bla bla bla 1. Aspect 1 2. Aspect 2 ---------- *Actual answer* ----------- (checkmark e…

Oh god I _hate_ this. Does anyone have any custom instructions to shut this thing off. The only thing that worked for me is to ask the model to be terse. But that causes the main answer part to be terse too, which sucks sometimes.

Chatgpt has a setting where you can set the tone to robotic

Re: Improved Gemini 2.5 Flash and Flash-Lite

#123

Post by fzimmermann89 »

The switch by Artificial Analysis from per-token-cost to per-benchmark-cost shows some effect! Its nice that labs are now trying to optimize what I actually have to pay to get an answer - It always annoys me to have to pay for all the senseless rambling of the less-capable reasoning models.

Re: Improved Gemini 2.5 Flash and Flash-Lite

#124

Post by tclancy »

canadiantim wrote: post #83
simonw wrote: post #74

I added support to these models to my llm-gemini plugin, so you can run them like this (using uvx so no need to install anything first): export LLM_GEMINI_KEY='...' uvx --isolated --with llm-gemini llm -m gemini-flash-lite-latest 'An epic poem about frogs at war with ducks' Release notes: https://github.com/simonw/llm-gemini/releases/tag/0.26 Pelicans: https://github.com/simonw/llm-gemini/issues/104#issuecomment...

Who wins in the end? the frogs? the ducks? or the pelicans?

I heard the dragon took the pole, but it may have been wind-aided.

Re: Improved Gemini 2.5 Flash and Flash-Lite

#125

Post by nharada »

newfocogi wrote: post #3

Non-AI Summary: Both models have improved intelligence on Artificial Analysis index with lower end-to-end response time. Also 24% to 50% improved output token efficiency (resulting in lower cost). Gemini 2.5 Flash-Lite improvements include better instruction following, reduced verbosity, stronger multimodal & translation capabilities. Gemini 2.5 Flash improvements include better agentic tool use and more token-effici…

I'm stealing "Non-AI Summary"

Re: Improved Gemini 2.5 Flash and Flash-Lite

#127

Post by kridsdale1 »

breakingcups wrote: post #84
jjani wrote: post #42

Earlier quoted context omitted.

Can't agree with that. Gemini doesn't lead just on price/performance - ironically it's the best "normie" model most of the time, despite it's lack of popularity with them until very recent. It's bad at agentic stuff, especially coding. Incomparably so compared to Claude and now GPT-5. But if it's just about asking it random stuff, and especially going on for very long in the same conversation - which non-tech users h…

My pet theory without any strong foundation is because OpenAI and Anthropic have trained their models really hard to fit the sycophantic mold of: =============================== Got it — *compliment on the info you've shared*, *informal summary of task*. *Another compliment*, but *downside of question*. ---------- (relevant emoji) Bla bla bla 1. Aspect 1 2. Aspect 2 ---------- *Actual answer* ----------- (checkmark e…

I suspect this has emerged organically from the user given RLHF via thumb voting in the apps. People LIKE being treated this way so the model converges in that direction.

Same as social media converging to rage bait. The user base LIKES it subconsciously. Nobody at the companies explicitly added that to content recommendation model training. I know, for the latter, as I was there.

Re: Improved Gemini 2.5 Flash and Flash-Lite

#128

Post by GaggiX »

dcchambers wrote: post #79

Why do all of these model providers have such issues naming/versioning them? Why even use a version number (2.5) if you aren't going to change it when you update the model? This industry desperately needs a Steve Jobs to bring some sanity to the marketing.

The version number is about the architecture of the model, the date is just about the last weights of the model.

Re: Improved Gemini 2.5 Flash and Flash-Lite

#129

Post by kridsdale1 »

typpilol wrote: post #121
breakingcups wrote: post #84

Earlier quoted context omitted.

My pet theory without any strong foundation is because OpenAI and Anthropic have trained their models really hard to fit the sycophantic mold of: =============================== Got it — *compliment on the info you've shared*, *informal summary of task*. *Another compliment*, but *downside of question*. ---------- (relevant emoji) Bla bla bla 1. Aspect 1 2. Aspect 2 ---------- *Actual answer* ----------- (checkmark e…

Anthropic also injects these long conversation reminders that are paragraph upon paragraphs about safety and what not to do. People have said it destroys the intelligence mid convo

Yes, but that’s their brand.

Re: Improved Gemini 2.5 Flash and Flash-Lite

#130

Post by m101 »

davidmckayv wrote: post #53

This really captures something I've been experiencing with Gemini lately. The models are genuinely capable when they work properly, but there's this persistent truncation issue that makes them unreliable in practice. I've been running into it consistently, responses that just stop mid-sentence, not because of token limits or content filters, but what appears to be a bug in how the model signals completion. It's been…

I wonder if this is because a memory cap was reached at that output token. Perhaps they route conversations to different hardware depending on how long they expect it to be.
Post reply on HN