Hi everyone,

I’m working on a college project (real-time waste segregation using YOLO + an agent layer) and I’ve run into a frustrating situation with Roboflow. I’d really appreciate some guidance from people who’ve dealt with this before.

Context

• Total dataset: ~20,000 images
• Annotated: ~11,111 images
• Annotation type: segmentation masks (SAM-assisted)
• Classes: ~18 categories (waste types like plastic_container, organic_waste, etc.)
• Goal: Export dataset → train YOLO locally (RTX 4050)

Problem

I used Roboflow’s AI annotation heavily and ended up with:

• ~81.26 credits used (shown in dashboard)
• Currently on Free/Public plan
• I purchased 1 prepaid credit

Now when I try to create a dataset version (even without augmentation), I get: “You have reached your monthly usage limit… your credits will reset on April 1”

So basically:

• I have 1 prepaid credit
• But I still cannot create a version
• No option for “flex billing” (free plan limitation)

Questions

1. Will I be able to create a version and export normally after April 1 reset?
2. Do prepaid credits NOT work if monthly free credits are exhausted?
3. Has anyone successfully:

• Exported dataset via API without creating a version?
• Bypassed this restriction somehow?