Skip to main content
The VLM Run API is a unified platform for production-ready multimodal AI. Use it to extract structured data from documents, images, videos, and audio — or run complex multi-step workflows with visual agents.
  • Base URL: https://api.vlm.run/v1
  • Authentication: Authorization: Bearer <VLMRUN_API_KEY>
  • Models Supported:
    • Requests: vlm-1
    • Agent Executions / Chat Completions: vlmrun-orion-1:auto, vlmrun-orion-1:fast, vlmrun-orion-1:pro
See Ways to Use VLM Run for a side-by-side comparison of Requests, Executions, Chat Completions, and the Chat UI.
Access your API keys in our dashboard.

​
Structured Extraction

Use the Generate endpoints to extract structured JSON from images, documents, audio, and video. 
from pathlib import Path
from vlmrun.client import VLMRun
from vlmrun.client.types import PredictionResponse

# Initialize the client
client = VLMRun(api_key="<VLMRUN_API_KEY>")

# Document -> JSON
response: PredictionResponse = client.document.generate(
    file=Path("path/to/document.pdf"),
    model="vlm-1",
    domain="document.invoice",
)

​
Chat Completions & Agent Executions

Use the Chat Completions endpoint for interactive multi-modal conversations, or the Agent Executions endpoint for batch execution workflows. 
from vlmrun.client import VLMRun

# Initialize the VLM Run client
client = VLMRun(api_key="<VLMRUN_API_KEY>")

# Create a chat completion
response = client.agent.completions.create(
    model="vlmrun-orion-1:auto",
    messages=[
        {
            "role": "user",
            "content": [
                {"type": "text", "text": "What do you see in this image?" },
                {"type": "image_url", "image_url": {"url": "https://example.com/image.jpg"}}
            ]
        }
    ],
    max_tokens=1000
)