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

Gemini 2.5 Flash

developers.googleblog.com

Post reply on HN
Print view View original

501–510 of 582 posts

Re: Gemini 2.5 Flash

#501

Post by vladmdgolam »

blueyes wrote: post #229
jeeeb wrote: post #192

Earlier quoted context omitted.

After comparing Gemini Pro and Claude Sonnet 3.7 coding answers side by side a few times, I decided to cancel my Anthropic subscription and just stick to Gemini.

One of the main advantages Anthropic currently has over Google is the tooling that comes with Claude Code. It may not generate better code, and it has a lower complexity ceiling, but it can automatically find and search files, and figure out how to fix a syntax error fast.

There are at least 10 projects currently aiming to recreate Claude Code, but for Gemini. For example, geminicodes.co by NotebookLM’s founding PM Raiza Martin

Re: Gemini 2.5 Flash

#502

Post by renewiltord »

profsummergig wrote: post #499

I tried this prompt in both Gemini 2.5 Pro, and in ChatGPT. "Draw me a timeline of all the dynasties of China. Imagine a horizontal line. Start from the leftmost point and draw segments for the start and end of each dynasty. For periods where multiple dynasties existed simultaneously draw parallel lines or boxes to represent the concurrent rule." Gemini's response: "I'm just a language model, so I can't help you with…

All the communities where people think LLMs are junk love Gemini. Makes me sceptical that the enthusiasm is useful signal.

I found the full 2.0 useful for transcription of images. Very good OCR. But not a good assistant. Stalls often and once it has, loses context easily.

Re: Gemini 2.5 Flash

#503

Post by syndeo »

brookst wrote: post #442
snthpy wrote: post #397

Earlier quoted context omitted.

Interesting that Americans appear to hold their AI models to a higher standard than their politicians.

Different Americans.

Lots of folks in tech have different opinions than you may expect. Many will either keep quiet or play along to keep the peace/team cohesion, but you really never know if they actually agree deep down.

Their career, livelihoods, ability to support their families, etc. are ultimately on the line, so they'll pay lip service if they have to. Consider it part of the job at that point; personal beliefs are often left at the door.

Re: Gemini 2.5 Flash

#504

Post by asim »

I just wish the whole industry would stop using terms like thinking and reasoning. This is not what's happening. If we could come up with more appropriate terms that don't treat these models like they're human then we'd be in a much better place. That aside, it's cool to see the advancement of Google's offering.

Re: Gemini 2.5 Flash

#505

Post by dbbk »

asim wrote: post #504

I just wish the whole industry would stop using terms like thinking and reasoning. This is not what's happening. If we could come up with more appropriate terms that don't treat these models like they're human then we'd be in a much better place. That aside, it's cool to see the advancement of Google's offering.

Thinking perhaps, but why not reasoning?

Re: Gemini 2.5 Flash

#506

Post by ramesh31 »

darksaints wrote: post #492
ramesh31 wrote: post #485

Earlier quoted context omitted.

>It's probably great for lots of things but it doesn't seem very good for recent news. You are missing the point here. The LLM is just the “reasoning engine” for agents now. Its corpus of facts are meaningless, and shouldn’t really be relied upon for anything. But in conjunction with a tool calling agentic process, with access to the web, what you described is now trivially doable. Single shot LLM usage is not really…

That’s all fine and dandy, but if you google anything related to llm agents, you get 1000 answers to 100 questions, companies hawking their new “visual programming” agent composers, and a ton of videos of douchebags trying to be the Steve Jobs of AI. The concept I’m sure is fine, but execution of agentic anything is still the Wild Wild West and nobody knows what they’re really doing.

Indeed there is a mountain of snake oil out there at this point, but the underlying concepts are extremely simple, and can be implemented directly without frameworks.

I generally point people to Anthropic's seminal blog post on the topic: https://www.anthropic.com/engineering/building-effective-age...

Re: Gemini 2.5 Flash

#507

Post by mrinterweb »

blueyes wrote: post #229
jeeeb wrote: post #192

Earlier quoted context omitted.

After comparing Gemini Pro and Claude Sonnet 3.7 coding answers side by side a few times, I decided to cancel my Anthropic subscription and just stick to Gemini.

One of the main advantages Anthropic currently has over Google is the tooling that comes with Claude Code. It may not generate better code, and it has a lower complexity ceiling, but it can automatically find and search files, and figure out how to fix a syntax error fast.

I don't understand the appeal of investing in leaning and adapting your workflow to use an AI tool that is so tightly coupled to a single LLM provider, when there are other great AI tools available that are not locked to a single LLM provider. I would guess aider is the closest thing to claude code, but you can use pretty much any LLM.

The LLM field is moving so fast that what is the leading frontier model today, may not be the same tomorrow.

Pricing is another important consideration. https://aider.chat/docs/leaderboards/

Re: Gemini 2.5 Flash

#508

Post by jonas21 »

deanmoriarty wrote: post #178

Genuine naive question: when it comes to Google HN has generally a negative view of it (pick any random story on Chrome, ads, search, web, working at faang, etc. and this should be obvious from the comments), yet when it comes to AI there is a somewhat notable “cheering effect” for Google to win the AI race that goes beyond a conventional appreciation of a healthy competitive landscape, which may appear as a bit of a…

A lot of the negativity toward Google stems from the fact that they're the big, dominant player in search, ads, browsers, etc., rather than anything that they've done or any particular attribute of the company.

In AI, they're still seen as being behind OpenAI and others, so we don't see the same level of negativity.

Re: Gemini 2.5 Flash

#509

Post by thegeomaster »

renewiltord wrote: post #502
profsummergig wrote: post #499

I tried this prompt in both Gemini 2.5 Pro, and in ChatGPT. "Draw me a timeline of all the dynasties of China. Imagine a horizontal line. Start from the leftmost point and draw segments for the start and end of each dynasty. For periods where multiple dynasties existed simultaneously draw parallel lines or boxes to represent the concurrent rule." Gemini's response: "I'm just a language model, so I can't help you with…

All the communities where people think LLMs are junk love Gemini. Makes me sceptical that the enthusiasm is useful signal. I found the full 2.0 useful for transcription of images. Very good OCR. But not a good assistant. Stalls often and once it has, loses context easily.

Is it possible that a community of people who are constantly pushing LLMs to their limits would be most aware of their limitations, and so more inclined to think they are junk?

In terms of business utility, Google has had great releases ever since the 2.0 family. Their models have never missed some mark --- either a good price/performance ratio, insane speeds, novel modalities (they still have the only API for autoregressive image generation atm), state-of-the-art long context support and coding ability (Gemini 2.5), etc.

However, most average users are using these models through a chat-like UI, or via generic tools like Cursor, which don't really optimize their pipelines to capture the strengths of different models. This way, it's very difficult to judge a model objectively. Just look at the obscene sycophancy exhibited by chatgpt-4o-latest and how it lifted LMArena scores.

Re: Gemini 2.5 Flash

#510

Post by aerhardt »

I am only on OpenAI because they have a native Mac app. Call me old-school but my preferred workflow is still for the most part just asking narrow questions and copying-pasting back and forth. I've been playing with Junie (Jetbrain's AI agent) for a couple of days, but I still don't trust agents to run loose in my codebase for any sizeable amount of work.

Does anyone know if Google is planning native apps? Or any wrapping interfaces that work well on a Mac?

Post reply on HN