GET /queue GET /queue GET /api/queue
Retrieves the current state of the execution queue, including running and pending prompts.
Response Array of currently executing prompts. Each item is an array containing: [number, prompt_id, workflow, extra_data, outputs_to_execute]
Array of prompts waiting to be executed. Each item follows the same structure as queue_running.
Example Request curl http://127.0.0.1:8188/queue
Example Response { "queue_running" : [ [ 42 , "a1b2c3d4-e5f6-7890-abcd-ef1234567890" , { "3" : { "class_type" : "KSampler" , "inputs" : { "seed" : 12345 , "steps" : 20 } } }, { "client_id" : "my-client" , "create_time" : 1709596800000 }, [ "9" ] ] ], "queue_pending" : [ [ 43 , "b2c3d4e5-f6a7-8901-bcde-f23456789012" , { "3" : { "class_type" : "KSampler" , "inputs" : { "seed" : 67890 , "steps" : 25 } } }, { "client_id" : "another-client" , "create_time" : 1709596820000 }, [ "9" ] ] ] }
Python Example import json import urllib.request def get_queue (): """Get current queue status.""" with urllib.request.urlopen( "http://127.0.0.1:8188/queue" ) as response: return json.loads(response.read()) queue = get_queue() print ( f "Running: { len (queue[ 'queue_running' ]) } prompts" ) print ( f "Pending: { len (queue[ 'queue_pending' ]) } prompts" ) for item in queue[ 'queue_pending' ]: number, prompt_id, workflow, extra_data, outputs = item print ( f " Prompt { prompt_id } at position { number } " )
POST /queue POST /queue POST /api/queue
Manages the execution queue by clearing all pending items or deleting specific prompts.
Request Body When true, removes all pending prompts from the queue. Does not affect currently running prompts.
Array of prompt IDs to remove from the queue.
Response Returns 200 OK on success.
Example: Clear Queue curl -X POST http://127.0.0.1:8188/queue \ -H "Content-Type: application/json" \ -d '{"clear": true}'
Example: Delete Specific Prompts curl -X POST http://127.0.0.1:8188/queue \ -H "Content-Type: application/json" \ -d '{ "delete": [ "a1b2c3d4-e5f6-7890-abcd-ef1234567890", "b2c3d4e5-f6a7-8901-bcde-f23456789012" ] }'
Python Example import json import urllib.request def clear_queue (): """Clear all pending prompts from the queue.""" data = json.dumps({ "clear" : True }).encode( 'utf-8' ) req = urllib.request.Request( "http://127.0.0.1:8188/queue" , data = data, headers = { 'Content-Type' : 'application/json' } ) urllib.request.urlopen(req) print ( "Queue cleared" ) def delete_from_queue ( prompt_ids ): """Delete specific prompts from the queue. Args: prompt_ids: List of prompt IDs to delete """ data = json.dumps({ "delete" : prompt_ids}).encode( 'utf-8' ) req = urllib.request.Request( "http://127.0.0.1:8188/queue" , data = data, headers = { 'Content-Type' : 'application/json' } ) urllib.request.urlopen(req) print ( f "Deleted { len (prompt_ids) } prompts" ) # Usage clear_queue() # Or delete specific prompts delete_from_queue([ "a1b2c3d4-e5f6-7890-abcd-ef1234567890" , "b2c3d4e5-f6a7-8901-bcde-f23456789012" ])
GET /prompt GET /prompt GET /api/prompt
Retrieves queue information including the number of remaining tasks.
Response Execution information object. Show exec_info properties
Number of tasks remaining in the queue (both running and pending).
Example Request curl http://127.0.0.1:8188/prompt
Example Response { "exec_info" : { "queue_remaining" : 5 } }
POST /interrupt POST /interrupt POST /api/interrupt
Interrupts the currently executing prompt or a specific prompt by ID.
Request Body Optional prompt ID to interrupt. If provided, only interrupts if this specific prompt is currently running. If omitted, performs a global interrupt on any running prompt.
Response Returns 200 OK on success.
Example: Global Interrupt curl -X POST http://127.0.0.1:8188/interrupt \ -H "Content-Type: application/json" \ -d '{}'
Example: Targeted Interrupt curl -X POST http://127.0.0.1:8188/interrupt \ -H "Content-Type: application/json" \ -d '{"prompt_id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890"}'
Python Example import json import urllib.request def interrupt_execution ( prompt_id = None ): """Interrupt prompt execution. Args: prompt_id: Optional specific prompt to interrupt. If None, interrupts any running prompt. """ payload = {} if prompt_id: payload[ "prompt_id" ] = prompt_id data = json.dumps(payload).encode( 'utf-8' ) req = urllib.request.Request( "http://127.0.0.1:8188/interrupt" , data = data, headers = { 'Content-Type' : 'application/json' } ) urllib.request.urlopen(req) if prompt_id: print ( f "Interrupted prompt { prompt_id } " ) else : print ( "Sent global interrupt" ) # Global interrupt interrupt_execution() # Targeted interrupt interrupt_execution( "a1b2c3d4-e5f6-7890-abcd-ef1234567890" )
POST /free POST /free POST /api/free
Controls memory management by unloading models or freeing memory.
Request Body When true, unloads all models from memory after the current execution completes.
When true, triggers memory cleanup after the current execution completes.
Response Returns 200 OK on success.
Example Request curl -X POST http://127.0.0.1:8188/free \ -H "Content-Type: application/json" \ -d '{ "unload_models": true, "free_memory": true }'
Python Example import json import urllib.request def free_memory ( unload_models = False , free_memory = False ): """Free memory and/or unload models. Args: unload_models: Whether to unload models from memory free_memory: Whether to free memory """ payload = {} if unload_models: payload[ "unload_models" ] = True if free_memory: payload[ "free_memory" ] = True data = json.dumps(payload).encode( 'utf-8' ) req = urllib.request.Request( "http://127.0.0.1:8188/free" , data = data, headers = { 'Content-Type' : 'application/json' } ) urllib.request.urlopen(req) print ( "Memory management flags set" ) # Unload models and free memory free_memory( unload_models = True , free_memory = True )
Complete Queue Management Example import json import urllib.request import time class QueueManager : def __init__ ( self , server_address = "127.0.0.1:8188" ): self .server = server_address self .base_url = f "http:// { server_address } " def get_queue_status ( self ): """Get current queue status.""" with urllib.request.urlopen( f " { self .base_url } /queue" ) as response: return json.loads(response.read()) def get_queue_remaining ( self ): """Get number of remaining tasks.""" with urllib.request.urlopen( f " { self .base_url } /prompt" ) as response: data = json.loads(response.read()) return data[ 'exec_info' ][ 'queue_remaining' ] def clear_queue ( self ): """Clear all pending prompts.""" self ._post_queue({ "clear" : True }) def delete_prompts ( self , prompt_ids ): """Delete specific prompts from queue.""" self ._post_queue({ "delete" : prompt_ids}) def interrupt ( self , prompt_id = None ): """Interrupt execution.""" payload = { "prompt_id" : prompt_id} if prompt_id else {} self ._post( f " { self .base_url } /interrupt" , payload) def free_memory ( self , unload_models = True ): """Free memory and unload models.""" self ._post( f " { self .base_url } /free" , { "unload_models" : unload_models, "free_memory" : True }) def wait_for_queue_empty ( self , poll_interval = 1.0 ): """Wait until queue is empty.""" while True : remaining = self .get_queue_remaining() if remaining == 0 : break print ( f "Waiting... { remaining } tasks remaining" ) time.sleep(poll_interval) print ( "Queue is empty" ) def _post ( self , url , payload ): """Helper to make POST requests.""" data = json.dumps(payload).encode( 'utf-8' ) req = urllib.request.Request( url, data = data, headers = { 'Content-Type' : 'application/json' } ) urllib.request.urlopen(req) def _post_queue ( self , payload ): """Post to queue endpoint.""" self ._post( f " { self .base_url } /queue" , payload) # Usage manager = QueueManager() # Check queue status status = manager.get_queue_status() print ( f "Running: { len (status[ 'queue_running' ]) } " ) print ( f "Pending: { len (status[ 'queue_pending' ]) } " ) # Wait for completion manager.wait_for_queue_empty() # Clean up manager.free_memory( unload_models = True )
Notes Queue operations affect pending prompts. Currently running prompts cannot be removed from the queue, but can be interrupted using /interrupt.
Clearing the queue or interrupting execution will cause any affected prompts to not complete. Make sure to handle this in your application logic.
Use the WebSocket API to receive real-time queue status updates instead of polling the REST endpoints.
See Also