Connect YOLO and language models in a visual workflow
Use YOLO to decide which images need a closer look, ask a vision-language model to explain them, or collect selected images in a dataset for review.
Connect images, models, and actions on a visual canvas
Agents is a visual workflow builder for computer vision. Add blocks, connect an output to the next step, and branch from one condition to several actions — for example, save an image and notify Slack from the same gate.
- YOLO, deployments, and language models: Run an official YOLO26 model, a trained workspace model, or a ready Platform deployment, then send matching images to a vision-language model.
- Gates that filter detections: Continue when a class count crosses a threshold, or when confidence falls in a range you choose.
- Inspect the Python: Open the generated workflow code at any time, and save a draft before or after you run it.
YOLO decides. A vision-language model explains.
Run YOLO first, then spend language-model calls only on the images that match. If no person is detected, the condition finishes without running the description branch.
- Start with YOLO26 nano: The shared trial runs an official YOLO26 nano model so you can try a small workflow in seconds.
- Your prompt, your provider: Choose a vision-language model and prompt. Add its key in Settings → API Keys; team workspaces use a key added by the workspace owner.
- Your trained models too: Select a dedicated execution deployment to run a trained workspace model or a larger workflow.
Need a model for your workflow?
Train YOLO26 on cloud GPUs, then connect it in Agents.
Explore industry solutions
See how teams apply Ultralytics computer vision across production environments.

Computer vision for aerial imagery

Computer vision in aerospace

Computer vision in security

Computer vision in robotics

Computer vision in logistics

Computer vision in retail

Computer vision in healthcare

Computer vision in manufacturing

Computer vision in automotive

Computer vision in agriculture

Computer vision for aerial imagery

Computer vision in aerospace

Computer vision in security

Computer vision in robotics

Computer vision in logistics

Computer vision in retail

Computer vision in healthcare

Computer vision in manufacturing

Computer vision in automotive

Computer vision in agriculture

Computer vision for aerial imagery

Computer vision in aerospace

Computer vision in security

Computer vision in robotics

Computer vision in logistics

Computer vision in retail

Computer vision in healthcare

Computer vision in manufacturing

Computer vision in automotive

Computer vision in agriculture
Enterprise-grade security, certified
Independently certified and audited, so your data stays protected in transit and at rest.
Run your first visual workflow
Start from a template on Ultralytics Platform. Connect YOLO, language models, and actions, then click Run.




