CogVLM¶
Class: CogVLMBlockV1
Source: inference.core.workflows.core_steps.models.foundation.cog_vlm.v1.CogVLMBlockV1
CogVLM reached End Of Life
Due to dependencies conflicts with newer models and security vulnerabilities discovered in transformers
library patched in the versions of library incompatible with the model we announced End Of Life for CogVLM
support in inference, effective since release 0.38.0.
We are leaving this block in ecosystem until release 0.42.0 for clients to get informed about change that
was introduced.
Starting as of now, all Workflows using the block stop being functional (runtime error will be raised),
after inference release 0.42.0 - this block will be removed and Execution Engine will raise compilation
error seeing the block in Workflow definition.
Ask a question to CogVLM, an open source vision-language model.
This model requires a GPU and can only be run on self-hosted devices, and is not available on the Roboflow Hosted API.
This model was previously part of the LMM block.
Type identifier¶
Use the following identifier in step "type" field: roboflow_core/cog_vlm@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 |
Text prompt to the CogVLM model. | ✅ |
json_output_format |
Dict[str, str] |
Holds dictionary that maps name of requested output field into its description. | ❌ |
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 CogVLM in version v1.
- inputs:
Stability AI Outpainting,OpenAI-Compatible LLM,Morphological Transformation,Contrast Enhancement,Crop Visualization,Camera Focus,Blur Visualization,Image Preprocessing,Corner Visualization,Ellipse Visualization,Mask Visualization,Stability AI Image Generation,Qwen-VL,Object Detection Model,Heatmap Visualization,Stitch OCR Detections,Google Gemma API,Roboflow Vision Events,Image Slicer,Qwen 3.5 API,Trace Visualization,Background Color Visualization,Slack Notification,Anthropic Claude,Qwen 3.6 API,OpenAI,Webhook Sink,Email Notification,Color Visualization,Bounding Box Visualization,Keypoint Visualization,Model Comparison Visualization,Google Gemma,Relative Static Crop,CogVLM,Llama 3.2 Vision,Polygon Zone Visualization,Qwen3.5-VL,Dynamic Crop,Camera Focus,Polygon Visualization,QR Code Generator,Google Vision OCR,GLM-OCR,Stitch Images,Google Gemini,Llama 3.2 Vision,OpenRouter,Twilio SMS Notification,Image Blur,Model Monitoring Inference Aggregator,Clip Comparison,Anthropic Claude,Image Slicer,Depth Estimation,OpenAI,Multi-Label Classification Model,Google Gemini,Classification Label Visualization,Pixelate Visualization,EasyOCR,SIFT,Florence-2 Model,MoonshotAI Kimi,Contrast Equalization,Image Threshold,Instance Segmentation Model,MoonshotAI Kimi,Dot Visualization,Polygon Visualization,Background Subtraction,Roboflow Dataset Upload,Anthropic Claude,Halo Visualization,Stability AI Inpainting,Roboflow Custom Metadata,Florence-2 Model,Label Visualization,Local File Sink,Icon Visualization,Single-Label Classification Model,Image Contours,OpenAI,Absolute Static Crop,Google Gemini,Grid Visualization,VLM As Classifier,Camera Calibration,Halo Visualization,Email Notification,OpenAI,LMM,LMM For Classification,Text Display,Image Convert Grayscale,Reference Path Visualization,Circle Visualization,Line Counter Visualization,Stitch OCR Detections,OCR Model,Keypoint Detection Model,SIFT Comparison,VLM As Detector,Morphological Transformation,Twilio SMS/MMS Notification,Roboflow Dataset Upload,CSV Formatter,S3 Sink,Triangle Visualization,Perspective Correction - outputs:
Stability AI Outpainting,CLIP Embedding Model,Delta Filter,SAM 3,Motion Detection,Contrast Enhancement,Image Preprocessing,Corner Visualization,Ellipse Visualization,Cosine Similarity,Object Detection Model,Heatmap Visualization,VLM As Classifier,OpenAI,Byte Tracker,Keypoint Visualization,Detections Consensus,First Non Empty Or Default,JSON Parser,Single-Label Classification Model,Dynamic Crop,Polygon Visualization,QR Code Generator,Stitch Images,OpenRouter,Semantic Segmentation Model,Dynamic Zone,Time in Zone,Segment Anything 2 Model,Buffer,EasyOCR,Contrast Equalization,Image Threshold,Anthropic Claude,Halo Visualization,Keypoint Detection Model,Detection Offset,Image Contours,OpenAI,SAM2 Video Tracker,SAM 3,Grid Visualization,VLM As Detector,Object Detection Model,Image Convert Grayscale,Reference Path Visualization,Stitch OCR Detections,Dominant Color,SIFT Comparison,Identify Changes,Roboflow Dataset Upload,CSV Formatter,Object Detection Model,Mask Visualization,Stability AI Image Generation,Qwen-VL,Stitch OCR Detections,Velocity,Qwen3.5,Perception Encoder Embedding Model,Bounding Box Visualization,Google Gemma,Detection Event Log,CogVLM,Bounding Rectangle,Instance Segmentation Model,SAM 3,Llama 3.2 Vision,Inner Workflow,SmolVLM2,Detections Stabilizer,Anthropic Claude,OpenAI,Depth Estimation,Image Slicer,Gaze Detection,PTZ Tracking (ONVIF),MoonshotAI Kimi,MoonshotAI Kimi,Line Counter,Keypoint Detection Model,Per-Class Confidence Filter,Semantic Segmentation Model,QR Code Detection,Detections Merge,Absolute Static Crop,Google Gemini,VLM As Classifier,Email Notification,OpenAI,Pixel Color Count,Rate Limiter,LMM For Classification,Text Display,Mask Area Measurement,Circle Visualization,Detections Combine,Image Stack,Mask Edge Snap,Perspective Correction,Multi-Label Classification Model,Camera Focus,Seg Preview,Roboflow Vision Events,Trace Visualization,OC-SORT Tracker,Time in Zone,Email Notification,Model Comparison Visualization,YOLO-World Model,Polygon Zone Visualization,Byte Tracker,GLM-OCR,Model Monitoring Inference Aggregator,Clip Comparison,Image Blur,Cache Get,Detections Stitch,Detections List Roll-Up,Instance Segmentation Model,Google Gemini,Pixelate Visualization,SIFT,Instance Segmentation Model,Polygon Visualization,Data Aggregator,Qwen2.5-VL,Roboflow Custom Metadata,Florence-2 Model,Local File Sink,Property Definition,Icon Visualization,Single-Label Classification Model,Detections Filter,ByteTrack Tracker,Barcode Detection,Size Measurement,Multi-Label Classification Model,LMM,Detections Transformation,Keypoint Detection Model,S3 Sink,Cache Set,BoT-SORT Tracker,SIFT Comparison,OpenAI-Compatible LLM,Morphological Transformation,Qwen3-VL,Identify Outliers,Crop Visualization,Blur Visualization,Google Gemma API,Image Slicer,Qwen 3.5 API,Path Deviation,Background Color Visualization,Slack Notification,Anthropic Claude,Dimension Collapse,Qwen 3.6 API,Webhook Sink,Color Visualization,Relative Static Crop,Llama 3.2 Vision,Path Deviation,Instance Segmentation Model,Qwen3.5-VL,Expression,Google Vision OCR,Camera Focus,Google Gemini,Distance Measurement,Single-Label Classification Model,SORT Tracker,Twilio SMS Notification,Moondream2,Multi-Label Classification Model,Template Matching,Classification Label Visualization,Florence-2 Model,Time in Zone,Dot Visualization,Background Subtraction,Roboflow Dataset Upload,Stability AI Inpainting,Line Counter,Detections Classes Replacement,Label Visualization,Overlap Filter,Halo Visualization,Camera Calibration,Clip Comparison,Continue If,Line Counter Visualization,Byte Tracker,OCR Model,VLM As Detector,Overlap Analysis,Twilio SMS/MMS Notification,Morphological Transformation,Triangle Visualization
Input and Output Bindings¶
The available connections depend on its binding kinds. Check what binding kinds
CogVLM in version v1 has.
Bindings
-
input
-
output
parent_id(parent_id): Identifier of parent for step output.root_parent_id(parent_id): Identifier of parent for step output.image(image_metadata): Dictionary with image metadata required by supervision.structured_output(dictionary): Dictionary.raw_output(string): String value.*(*): Equivalent of any element.
Example JSON definition of step CogVLM in version v1
{
"name": "<your_step_name_here>",
"type": "roboflow_core/cog_vlm@v1",
"images": "$inputs.image",
"prompt": "my prompt",
"json_output_format": {
"count": "number of cats in the picture"
}
}