Roadmap
Changelog

Have something to say?

Tell us how we could make the product more useful to you.

Feedback: First time use

Wanted to share some feedback from my recent session: Generation time: The time to generate results felt very long, especially given that the initial SVG I uploaded wasn’t particularly complex. Even if some operations are heavy, a clearer sense of progress or expected duration would help set expectations. Enhanced text UX: When I use the “Enhanced” text feature, the enhanced output is added below my original text in the textarea. It would be really helpful to have explicit “Discard” and “Accept & Replace” actions. That way I could choose to replace the current textarea value with the enhanced version, instead of ending up with both versions together. I just wasn’t sure it was pulling through to the next step. AI Generation output: In one of my AI generations, the result included literal “```xml” text in the output, which looks like leftover formatting rather than part of the SVG. Overall really like the concept and visual direction, it looks great and was happy with the quality of the output (beside the floating xml text). Definitely something I would spend money on. Best of luck!

Sam about 1 month ago

1
🐛

Bug Reports

Generation was Interrupted

Image was generating, it’s taking its time. I go to settings and then out of the project and the generation stopped, i imagined it would continue in the background but it didn’t. It would be nice either if it did continue in the background or and alert told me to be careful of stepping out of the project because the generation will stop. Maybe also when closing tab?

Beto Alanis about 1 month ago

1
🐛

Bug Reports

Prompt Enhancer

I tested several generations using the Gemini model instead of Auto mode, and the results were inconsistent. Without proper animations, the output was distorted. When the enhanced prompt option is enabled, it seems to modify what I originally intended to do. I am not sure what is happening, but when I typed the prompt directly without enhancements, the result was accurate. I think the AI prompt enhancer needs further improvement and training not to modify real intended scanario. Btw i’m happy with the last result below

Kasun Chanuka Fernando about 1 month ago

🐛

Bug Reports

User Guide Needed

When a newbie or non-technical user signs up and lands on the dashboard, they often have to figure out what to do on their own. As a developer, I know the drag-and-drop block builder is straightforward, but from a UX perspective it would be better to add a one-time guided tour (like Driver.js or tooltip based with steps) to explain what everything is, what happens next, and how AI credits work.

Kasun Chanuka Fernando about 1 month ago

💡

Feature Requests

General Feedback

Hi guys! I just wanna provide some quick feedback! First of all, really cool UI, I like it a lot. It would be perfect if you could add a few more features to each connection, but it's looking great so far. About my experience using it: Don't charge for server errors: I had a failed AI generation where the prompt threw a 'server error' but my credits were still used. That's not supposed to happen! From a user's perspective, a server error is 'your' error (or the provider's), and I have no control over that. You shouldn't "punish" the user for it—maybe add a check so it only "cashes out" the credits once the generation is actually successful? (should be an easy fix!) PNG to SVG converter: You guys should add a PNG -> SVG converter. For obvious reasons, it's way easier to find or generate a JPEG/PNG than creating an SVG from scratch. Maybe a plugin or a simple canvas where we can visually 'tweak' the SVG before the AI modifies it? (I know, I know... that’s probably a lot of work!) Resize the SVG window: The source window for the SVG is too tiny, it should be resizable. Also, the image inside should be adjustable—I couldn't do it with my MacBook trackpad. I uploaded an image but could barely see a fraction of it, so I wasn't sure if the platform cropped it or if the model was going to spend my credits on a cropped version. It gives you a lot of uncertainty. Preview / Pre-final results: It would be awesome to have a 'preview' to see if the design is going in the right direction. Maybe it could cost 'half credits' and then we can push it till the end? We all know AI doesn't always get it right on the first try and can create something totally different from what we planned. Play/Pause for animations: If we're talking about animated SVGs, it really needs a play/pause button. I tried creating a 'battery' that was supposed to fill up with colors, but I just got a black shape. It’s hard to tell if it's a prompt issue or a playback issue without a button to check the motion. The learning curve: Just a personal note, but with 30 credits I only got like 3 tries. I don’t think that’s enough to really 'get to the point' where you can use the tool properly. Think about the ‘learning curve’ but here is more about to calibrate the details the AI needs to receive to give you what you want.

Daniel Zuñiga about 2 months ago

🐛

Bug Reports