Get coding help from Gemini Code Assist – now for free | Hacker News Reader

Get coding help from Gemini Code Assist – now for free

blog.google

79 pointsxnx17 comments

I wasn't sure what this was built on top of. I've asked the model and told me this:

No, I'm not specifically "Gemini 2.0." Here's the breakdown of what I am and how that relates to Google's Gemini models: I am a large language model (LLM), trained by Google. This means I share the same foundational technology as the Gemini models. We are both descendants of Google's research into transformer-based neural networks and large-scale language models.

I am built on a variant of the models used in the Gemini family. While I don't have a specific version number like "Gemini 2.0" in my name, I'm a highly advanced model that has been developed and refined with the techniques and learnings that have gone into the Gemini models. This means i am a Mixture of Experts model.

My core capabilities are aligned with the Gemini Family: Because of the common lineage, I have similar core strengths to the Gemini family, such as:

* Understanding and responding to complex prompts.

* Generating creative content.

* Engaging in multi-turn conversations.

* Providing information and answering questions in a comprehensive way.

* Processing and understanding code.

The naming conventions for models are fluid and complex. Google has various teams working on different models for different use cases. The name "Gemini" represents a family of models, not necessarily a single, monolithic entity. Also, the naming used in the different versions may vary.

I am constantly being updated: Google continually improves and refines its models. My capabilities are always growing and evolving.

> We fine-tuned the Gemini 2.0 model for developers by analyzing and validating a large number of real-world coding use cases.

Gemini doesn't give a correct response when asked about what model it is. You can convince it to answer pretty much any version number. From the docs, it's a fine-tuned the Gemini 2.0 model for developers.

https://codeassist.google/products/business still shows this being Gemini 1.5 Pro. Maybe just not updated? Or is this a different product?

>This capability is powered by Google’s Gemini 1.5 Pro model.

I guess 1.5 Pro is too expensive to run this kind of product given it offers quite generous limits. More likely 2.0 Flash or Flash lite.

I really dislike how many of these titles say free when it isn't. It means that if it becomes popular, google will reduce the usage limits, then say "we never said it was free".

How do I test that??

I code in vim (or nano) with some basic sh scripts (certainly not bash). Then I would remotely query gemini via curl, or wget, something along those lines.

I could not see any pointers on some "web API" documentation, or maybe it is dramatically hidden in some javascript-only text.

And even if it is "somewhere", how do I register for a secret for this web API? I can bet, if it is possible, you must have a blink based browser, and care was probably taken at _NOT_ being compatible with noscript/basic (x)html browsers.

I hope I am wrong, if not, I can only despise those people as they are doing it clearly on purpose, and hope only for hardcore regulation on technical interfaces.

I hope not

JavaScript enabled browsers are the future and burdening engineers with forever supporting people with no script enabled should not be something we do

On HN, all humans, aka non AI bots, do understand the underlaying consequences and usually have a pertinent perspective on the technical cost of the web engines from the whatng cartel.

If you don't want to understand that, I cannot do much more.

come again?

>>How do I test that??

>>I code in vim (or nano) with some basic sh scripts (certainly not bash).

Though its definitely possible to configure vim/emacs to do anything. Im guessing you stray further away from the unlocking the true benefits of using LLMs with advanced IDEs like vscode.

This is not even editing in any real sense of the word. You are already way off, the true purpose and workflow of how code work is done using an LLM.

The CTA button on the left (Try at no cost) does not work (404).

You can find out more at https://codeassist.google/products/individual. The team is on this issue. Thanks for reporting!

Not sure if related. The register link on this post is also broken: https://blog.google/feed/kaggle-genai-intensive-course-2025/

That looks like a separate issue, but will investigate. Thank you!

Also Google:

* We force enabled Gemini on your Corporate Accounts, dont even try to turn it off suckers, haha

* A month later: We're raising prices because of the "benefits" of Gemini

Why is it so bad in this day and age of Cursor/Windsurf? I installed yesterday expecting I will loose my time since it’s made by Google (how far have they fallen, I remember being so excited about gmail in 2005). And ho boy, you can’t even drag and drop files from your ide in the chat to add them into the context. Basic features that we now take for granted are lacking. I uninstalled right away, don’t know why I fell for it once more.

Why is Google in that state? What happened?

They have to give their product for free and it’s so bad that even for free it seems nobody wants them. During that time you have company like Cursor that are super popular and charge for their service. Anthropic their servers are melting under the load and they charge top dollar for it.

If this came out 2 or 3 years ago, that would have been good.

Reply on news.ycombinator.com