Moondream2

Class: Moondream2BlockV1

Source: inference.core.workflows.core_steps.models.foundation.moondream2.v1.Moondream2BlockV1

This workflow block runs Moondream2, a multimodal vision-language model. You can use this block to run zero-shot object detection.

Type identifier

Use the following identifier in step "type" field: roboflow_core/moondream2@v1to add the block as as step in your workflow.

Properties

Name	Type	Description	Refs
name	str	Enter a unique identifier for this step..	❌
prompt	str	Optional text prompt to provide additional context to Moondream2. Otherwise it will just be None.	❌
model_version	str	The Moondream2 model to be used for inference..	✅

The Refs column marks possibility to parametrise the property with dynamic values available in workflow runtime. See Bindings for more info.

Available Connections

Compatible Blocks

Check what blocks you can connect to Moondream2 in version v1.

• inputs: Image Contours, Stability AI Inpainting, Mask Visualization, Corner Visualization, Keypoint Visualization, Line Counter Visualization, Reference Path Visualization, Circle Visualization, Absolute Static Crop, Image Blur, Relative Static Crop, Perspective Correction, Multi-Label Classification Model, Image Convert Grayscale, Instance Segmentation Model, Pixelate Visualization, Object Detection Model, Model Comparison Visualization, Triangle Visualization, Trace Visualization, Background Color Visualization, SIFT Comparison, Camera Calibration, Depth Estimation, Label Visualization, Image Threshold, Classification Label Visualization, Blur Visualization, Color Visualization, Image Preprocessing, Bounding Box Visualization, Keypoint Detection Model, Ellipse Visualization, Stability AI Image Generation, Polygon Zone Visualization, SIFT, Grid Visualization, Camera Focus, Stitch Images, Stability AI Outpainting, Polygon Visualization, Image Slicer, Image Slicer, Dynamic Crop, Single-Label Classification Model, Crop Visualization, Halo Visualization, Dot Visualization
• outputs: Line Counter, Detections Consensus, Corner Visualization, Time in Zone, Detections Stabilizer, Model Monitoring Inference Aggregator, Florence-2 Model, Circle Visualization, Line Counter, Stitch OCR Detections, Roboflow Dataset Upload, PTZ Tracking (ONVIF).md), Path Deviation, Byte Tracker, Detections Classes Replacement, Detections Merge, Byte Tracker, Pixelate Visualization, Model Comparison Visualization, Distance Measurement, Detections Stitch, Segment Anything 2 Model, Overlap Filter, Triangle Visualization, Velocity, Background Color Visualization, Roboflow Dataset Upload, Trace Visualization, Path Deviation, Florence-2 Model, Label Visualization, Blur Visualization, Color Visualization, Bounding Box Visualization, Detection Offset, Ellipse Visualization, Detections Transformation, Roboflow Custom Metadata, Size Measurement, Time in Zone, Byte Tracker, Dynamic Crop, Crop Visualization, Detections Filter, Perspective Correction, Dot Visualization

Input and Output Bindings

The available connections depend on its binding kinds. Check what binding kinds Moondream2 in version v1 has.

Bindings

• input

  • images (image): The image to infer on..
  • model_version (roboflow_model_id): The Moondream2 model to be used for inference..

• output

  • predictions (object_detection_prediction): Prediction with detected bounding boxes in form of sv.Detections(...) object.

Example JSON definition of step Moondream2 in version v1

{
    "name": "<your_step_name_here>",
    "type": "roboflow_core/moondream2@v1",
    "images": "$inputs.image",
    "prompt": "What is in this image?",
    "model_version": "moondream2/moondream2_2b_jul24"
}