Introducing the Gemini API CLI, now live in preview! 🚀 The easiest way for your agents to generate images, videos, music, text-to-speech or use managed agents. A fast and efficient way to get access to all the goodness of the Gemini API. 🧵👇

calling it Gemini API CLI while Gemini CLI was deprecated is confusing

I mean it is a CLI for using the Gemini API. So a bit of confusion is going to happen, hard to avoid. Open to suggestions though 😀

Yeah it makes sense. I think the confusion is more about how the other products were named for example, Gemini is the name of the Chat product but also the name of LLMs... and previously it was the name of the terminal coding harness (Gemini CLI). I like the move Google made to call the harness as Antigravity so we can distinguish it.

We are not the best at naming, that is no secret 😅 Totally agree, Antigravity is a great name. In this case the CLI is just a wrapper around the Gemini API and not a new product, so made sense to make the name the way it is.

@oviniciuslana How will it be different from Gemini API in Antigravity?

It will be added to that plugin shortly! Two different use-cases, will do a write-up explaining in more detail but our plugin will also cover this. Building an app that uses Gemini --> use our SDK Want to generate sound effects for a game, or a video for your website --> Gemini API CLI The CLI is a tool Antigravity (and any agent) can use to quickly generate media, audio etc. Our SDK is still building software that uses Gemini.

Some examples of when to use the Gemini API CLI: 🎮 - "Add a soundtrack for the boss fight in my game" 📊 - "Create a cover image for this week's newsletter" 📹 - "Make a 10s hero video for the landing page"

Launch blog: GitHub:

very slow,HTTP status: 503,"The server failed to process the request — retry after a short delay"

What command were you running? Will take a look!

Real question: does the CLI get the full long-context treatment, or is this the lightweight build?

Mind explaining more? This CLI is not a coding harness, it is a tool to be used by any coding agent to call the Gemini API. Whether that be to generate media, video, or audio, it is up to you.

@JackWoth98 Who do I have talk to begin speaking with someone from the team. I really wanna understand the decisions that are being made here when it comes to dev ex / product as a whole.

Feel free to send me a DM, happy to try and answer any questions you have.

I gotchu my friend!

Wtf I'm trying it now after this late night coffee. And def bridging to my pixel watch IMMEDIATELY

You guys got real creative with naming this one Any agent skills to go with it?

Couldnt care less

Looks great! Any plans to let us use our Gemini Advanced/subscription quota with the CLI instead of just API keys?

I thought it was Antigravity now?

What a banger video !! 🔥

give the failing test suite boss music

Can you guys include the option of subscription along api

How

Doesn't agy do all this already?

This is super useful

这个和 Antigrative CLI 的区别是什么？我感觉你们的 harness 种类似乎比较混乱😂

maybe pump some extra effort into AGY CLI now.

The file-path output is a nice detail: generate an asset, then hand that exact file to the next shell step. No need to extract a filename from a chatty success message.

Cheaper flash lite

Dude ask your gemini to make this naming correct it is confusing

Looked at the tts flags, saw --voice and --multi-speaker but nothing for language. Does it just follow whatever language the text is in?

So does this mean we can use the agy CLI to leverage a Google AI Pro/Ultra Subscription for Gemini LLM access (reasoning/coding), then use the Gemini API CLI for pay-as-you-go access to the other modality generators (Music, Image Gen/Edit, Video Gen/Edit, TTS, STT)?