User Guide

In Frame-to-Video, if you upload more than one image, two new settings show up. Custom Order: you choose which order your images get used in. Say you uploaded 3 photos — you pick which one goes first: the order you set yourself, alphabetical A to Z, or newest upload first. Image Processing Option: this decides how each photo is used in the video. Pick "Use Start frame only" and your photo just becomes the first second of the video, then Grok finishes the rest based on the text you wrote. Pick "Use Start and End frame" and you upload two photos: one becomes the start of the video, the other the end — Grok fills in what happens between them.

At the very top of the panel you'll see six buttons: Text to Video, Frame to Video, Ingredients to Video, Text to Image, Image to Image, Agent Automation. Only one can be selected at a time (shown in blue). Simply put: click the button matching what you want to create (video or image, and with what kind of input). All the settings below change based on the mode you pick.

This number sets how many prompts are processed together at the same time. Set it to 1 and prompts run one after another. Set it to 2 or 3 and more than one is processed at once, finishing faster. But a very high number can burn through your daily Grok quota faster. If unsure, start with 1 or 2.

Before moving from one finished prompt to the next, the extension waits a random amount of time between the two values you set here (in seconds, e.g. 20 to 30). This wait keeps Grok from being hit with requests too fast, one after another, and helps mimic normal usage. You usually don't need to change the defaults.

Here you type each thing you want to create, one per line. For example, if one line says "a red car" and the next says "a blue plane", these become two separate, independent prompts in the queue. If a single request spans multiple lines (it's long), leave a blank line between requests so the extension knows where one ends and the next begins. Instead of typing by hand, you can also upload a .txt file or an Excel/.csv file — the extension automatically turns the lines/cells inside into your prompt list.

If you've already created "characters" in Grok (say, a person/mascot that always looks the same), turning this on makes the extension automatically add that character to a prompt whenever its name appears in the text. For this to work, you first need to click "Scan Characters" so the extension can see your characters.

With "Create new" selected, each request works on its own, using the images you upload. With "Reuse" selected, the image/frame generated by the previous request is automatically used as the input for the next one — meaning generations chain together, one after another (useful for, say, consecutive frames of a story).

"Outputs per Prompt" sets how many different images/videos each request generates (e.g. set it to 2 and each request produces 2 variations). "Save to Folder" sets which subfolder, under your computer's Downloads folder, the downloaded files go into — leave it blank and they land in the default folder.

With "Auto Change File Name" on, downloaded files are saved with readable names (so it's clear which prompt they came from) instead of random ones. If "Auto Download" is off, generated images/videos aren't automatically downloaded to your computer — they only show up on the Grok page, and you download manually if you want.

After clicking "Run", this shows the live status of the requests you sent: which one is still processing, which finished, how many are done. Use the "Stop" button here if you want to halt a run.

"Report Bug": lets you tell us if something's wrong. "Clear Cache": try this if you see a strange error/warning on the Grok page — it clears and refreshes some of the page's cache and data. "Clear": wipes your written requests and the queue entirely, starting fresh. "Run": sends everything you've written to Grok and starts generating.