What it does
Igen makes and edits pictures through a bench of image models you choose between. Pick a tier and Igen picks the model, or pick the model yourself. The button shows the price before you spend it.
To change part of a picture, paint over it. White is the region that changes. Fill redraws it from a prompt; Remove takes it out with no prompt. Igen sends only the window around your paint, so a small fix on a large picture costs less.
Every picture is saved to your Hoq Storage with a record of how it was made written into the file itself: an iTXt chunk in PNG, XMP in JPEG and SVG.
In numbers
4pictures per request on most models
5aspect ratios on the default models: 1:1, 4:3, 3:4, 16:9, 9:16
1K, 2K, 4Koutput sizes on Gemini 3 Pro Image
Specifications
| Models on the bench | Ideogram v4 fast, FLUX.2 flex, Gemini 3 Pro Image, Seedream v5 Pro, Recraft v4 vector; FLUX Pro fill and Bria for edits. |
|---|---|
| Edit tools | Fill a painted region from a prompt, or remove it. The mask is strictly black and white. |
| Repeatable | Seeds are a per-model fact shown on each model; most text-to-image models do not offer one. |
| Metadata in the file | PNG iTXt, or XMP in JPEG and SVG, plus a sidecar record. |
| Who pays | Each picture is billed to your Hoq account at the price shown on the button. |
| Where pictures go | Your Hoq Storage. |
| Vector output | Recraft v4 returns real SVG files. |
Features
10 features, all shipped. Listed from the product's code and documentation; each row shows the day it was last checked.
| Text to image | Makes up to four pictures from a prompt. | |
| Tiers and models | Choose a quality tier or a specific model. | |
| Price on the button | Shows what a run costs before you start it. | |
| Vector | Makes an SVG through Recraft. |
| Fill | Redraws a painted region from a prompt. | |
| Remove | Erases a painted region with no prompt. | |
| Brush | Brush with undo by stroke, Tab to see the original, Alt+scroll to resize. |
| Saved to your drive | Keeps every picture in your Hoq Storage. | |
| Made-with record | Writes the prompt and model into the file. | |
| Retry | Runs a failed or finished request again. |
Works with
- storagePictures are saved in your drive.
- Hoq BoardBuilt to mount inside Hoq Board; not yet released there.
Your data
| Where it is kept | Your Hoq Storage. Prompts and source pictures are sent to the chosen model's provider. |
|---|---|
| Getting it out | Each picture is a file in your drive. |
| If you leave | Not documented. |
APIs and standards
- Web app
- Embeddable compose element for other Hoq apps (not yet released)
- Standard: PNG
- Standard: JPEG
- Standard: SVG
- Standard: XMP
Compared
| Checked 2026-11-01 | Hoq Igen | Midjourney | Canva Pro (AI tools) |
|---|---|---|---|
| Price | Pay per picture at the price shown; no subscription | $10, $30, $60 or $120 a month1 | US$120 a year for one person2 |
| Choice of model | Several providers' models per picture | Midjourney's own models1 | not published3 |
| Usage limits | None beyond your balance | About 3.3 to 60 fast GPU hours a month; extra at $4 an hour1 | Shared monthly AI allowance, 10x Free on Pro3 |
| Private by default | Pictures go to your own drive | Stealth mode on Pro and Mega only4 | not published3 |
| Commercial use for large businesses | not published | Pro or Mega needed above $1M a year gross5 | not published3 |
| Design layout and templates | No; layout builder not built | No1 | Yes, full design editor2 |
- https://docs.midjourney.com/hc/en-us/articles/27870484040333-Comparing-Midjourney-Plans
- https://canva.com/pricing/
- https://www.canva.com/help/ai-access/
- https://docs.midjourney.com/hc/en-us/articles/32019750070669-Stealth-Mode
- https://docs.midjourney.com/hc/en-us/articles/27870375276557-Using-Images-Videos-Commercially