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.. | ✅ |
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:
Stitch Images,Image Threshold,Email Notification,Corner Visualization,Image Blur,Ellipse Visualization,OpenAI,Roboflow Dataset Upload,Object Detection Model,Depth Estimation,Stitch OCR Detections,EasyOCR,Absolute Static Crop,Multi-Label Classification Model,CogVLM,Google Gemini,Stability AI Image Generation,Grid Visualization,Dynamic Crop,Image Slicer,Image Preprocessing,Relative Static Crop,SIFT,Morphological Transformation,Instance Segmentation Model,Line Counter Visualization,Trace Visualization,LMM For Classification,Halo Visualization,Dot Visualization,GLM-OCR,Model Monitoring Inference Aggregator,Roboflow Custom Metadata,Keypoint Detection Model,Pixelate Visualization,Circle Visualization,Image Convert Grayscale,Icon Visualization,QR Code Generator,S3 Sink,Semantic Segmentation Model,Keypoint Detection Model,Twilio SMS Notification,Halo Visualization,Camera Focus,Anthropic Claude,OCR Model,Polygon Visualization,Text Display,Reference Path Visualization,Instance Segmentation Model,Llama 3.2 Vision,CSV Formatter,Crop Visualization,Roboflow Dataset Upload,Mask Visualization,Heatmap Visualization,Webhook Sink,Label Visualization,Classification Label Visualization,Google Vision OCR,Florence-2 Model,Florence-2 Model,VLM As Detector,Polygon Zone Visualization,Stability AI Inpainting,Google Gemini,Perspective Correction,Camera Calibration,Anthropic Claude,OpenAI,OpenAI,Qwen3.5-VL,Background Color Visualization,Anthropic Claude,Email Notification,Background Subtraction,Contrast Equalization,SIFT Comparison,Multi-Label Classification Model,Keypoint Visualization,Stitch OCR Detections,LMM,Single-Label Classification Model,Color Visualization,Single-Label Classification Model,OpenAI,Object Detection Model,Roboflow Vision Events,Local File Sink,VLM As Classifier,Twilio SMS/MMS Notification,Triangle Visualization,Clip Comparison,Blur Visualization,Bounding Box Visualization,Camera Focus,Polygon Visualization,Google Gemini,Image Slicer,Image Contours,Model Comparison Visualization,Stability AI Outpainting,Slack Notification - outputs:
Corner Visualization,Overlap Filter,Ellipse Visualization,Perspective Correction,Roboflow Dataset Upload,Stitch OCR Detections,Time in Zone,PTZ Tracking (ONVIF),Time in Zone,Background Color Visualization,Dynamic Crop,Size Measurement,Velocity,Detections Combine,Trace Visualization,Detection Event Log,Time in Zone,Dot Visualization,ByteTrack Tracker,Line Counter,Model Monitoring Inference Aggregator,Path Deviation,Roboflow Custom Metadata,Stitch OCR Detections,Detections Filter,Segment Anything 2 Model,Pixelate Visualization,Detections Merge,Circle Visualization,Icon Visualization,Detections Classes Replacement,Detections Transformation,Detections Stabilizer,Camera Focus,Byte Tracker,Line Counter,OC-SORT Tracker,Detections Consensus,Color Visualization,Roboflow Vision Events,Crop Visualization,Roboflow Dataset Upload,Detection Offset,Mask Area Measurement,Detections Stitch,SORT Tracker,Triangle Visualization,Heatmap Visualization,Blur Visualization,Bounding Box Visualization,Byte Tracker,Byte Tracker,Label Visualization,Detections List Roll-Up,Distance Measurement,Florence-2 Model,Path Deviation,Florence-2 Model,Model Comparison 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..prompt(string): Optional text prompt to provide additional context to Moondream2..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": "my prompt",
"model_version": "moondream2/moondream2_2b_jul24"
}