Excited to share a new milestone today: we just passed $1M+ in credits given to researchers and academics who use Roboflow. This isn’t a future initiative we’re planning or one that starts today. We’ve been deeply invested in supporting new work by students, academics and researchers for years. You can browse over 8,000 papers that cite us at Roboflow and Research.
If you’re a student or researcher, apply for credits here!
Roboflow research plans
We offer two tiers of research credits for students and researchers in academia. The Training Credits plan is ideal if you are training models, whereas the Images plan is ideal if you are annotating or managing large datasets.
Training Credits:
- + 10k source images
- + 50k generated images
- + 22 training credits
Image Credits:
- + 90k source images
- + 450k generated images
- + 5 training credits
Both plans include:
- 15 team members
- 50 credits per month
- Enhanced image augmentations
- Model weights download
To apply for research credits:
- Create an account in Roboflow.
- Create a Public Workspace.
- Navigate to Settings > [Your Workspace] > Plan & Billing.
- Scroll to “Using Roboflow for research or education?” at the bottom of the page and click “Request Access”
By applying for an academic plan, all data in the workspace where you make your application will be made public on Roboflow Universe. You need an academic email address for approval and must be doing non-commercial work related to research or academia.
Get your work in front of thousands
We actively monitor the web for new scholarly mentions of Roboflow and add those papers to our Citations page daily. Make sure your research ends up there for others to find and reference by citing us.
Here are a few popular templates:
Cite Roboflow
Dwyer, B., Nelson, J., Hansen, T., et al. (2024). Roboflow (Version 1.0) [Software]. Available from https://roboflow.com. Computer vision.
Cite Roboflow 100
F. Ciaglia, F. Saverio Zuppichini, P. Guerrie, M. McQuade, and J. Solawetz, “Roboflow 100: A rich, multi-domain object detection benchmark,” 2022, arXiv:2211.13523.
Cite Roboflow 100-VL
P. Robicheaux, M. Popov, A. Madan, I. Robinson, J. Nelson, D. Ramanan, N. Peri, Roboflow100-vl: A multi-domain object detection benchmark for vision-language models, Roboflow (2025).
Cite RF-DETR
P. Robicheaux, M. Popov, I. Robinson, RF-DETR: SOTA Real-Time Object Detection Model, Roboflow (2025). https://github.com/roboflow/rf-detr
Cite a Roboflow Universe Dataset
Team Roboflow. "COCO 128 Dataset". Roboflow. https://universe.roboflow.com/team-roboflow/coco-128 (accessed DATE HERE)
Cite Our Blog
We’ve written over a thousand blog posts, covering a wide range of topics. Here’s an example of how you can cite one of our articles in your work:
Gallagher, J., & Skalski, P. (2023, September 27). GPT-4 with Vision: Complete Guide and Evaluation. Roboflow. https://blog.roboflow.com/gpt-4-vision/
Partner with Roboflow on research
In addition to supporting external researchers, we publish our own research, often in partnership with leading researchers and institutions. Earlier this month we presented Roboflow 100-VL at CVPR alongside Carnegie Mellon University.
Other notable partners include Intel on Roboflow 100, Microsoft Research on The 2nd Workshop on Computer Vision in the Wild, and more.
Working on something in this domain or have a new idea? Get in touch. We'd love to collaborate with you.
Apply today
We’ve been fortunate to support the work of over a million developers, many of which are students, researchers, professors, scientists and open source contributors. If you are in that group and could benefit from Roboflow’s platform, we encourage you to apply for research credits today. We can’t wait to see what you build.
Cite this Post
Use the following entry to cite this post in your research:
Jackson Prince. (Jun 20, 2025). Roboflow Powers AI Progress with $1M+ in Research Support. Roboflow Blog: https://blog.roboflow.com/roboflow-research/