vlmrun

What it is and what it does

VLM Run is the official Python SDK for interacting with the VLM Run API platform, a hosted service for vision-language models. It provides a client library to send images, documents, and text prompts to remote models and receive structured responses. The SDK wraps REST APIs for image generation (with domain-specific processing like invoice extraction), document analysis, and chat completions that are compatible with the OpenAI API format.

The package is designed for developers who want to integrate multimodal AI capabilities into Python applications without running models locally. It includes optional features for video processing, document handling, and OpenAI SDK compatibility, plus a CLI tool for interactive chat with skills (domain-specific expertise bundles). The SDK handles authentication, request formatting, and response parsing, reducing boilerplate when calling the platform's endpoints.

Use it for:

• Extract structured data from invoices, receipts, or other documents by sending images to domain-specific models via the image.generate endpoint.
• Build a Python application that uses OpenAI-compatible chat completions to interact with VLM Run's Orion model asynchronously.
• Process video frames or document PDFs using the optional video and document extras for batch analysis workflows.
• Create a CLI chatbot with domain expertise by uploading skills to the VLM Run platform or passing local skill directories.
• Integrate multimodal AI into data pipelines using pandas and image libraries alongside the SDK's image processing methods.

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Python SDK for the VLM Run API platform, providing access to vision-language models for image and document processing, chat completions, and CLI-based agent interactions.

Yes, if you have a VLM Run API account and need a Python client for their platform. The SDK is actively maintained, has low install friction, carries no known vulnerabilities, and uses a permissive license. The 15 runtime dependencies are substantial but manageable; consider using optional extras if you don't need all features. Not suitable if you need local model inference or don't have API credentials.

Install

vlmrun on PyPI

pip install vlmrun

uv add vlmrun

poetry add vlmrun

Installing vlmrun

Before you install

Low friction installation with a pure-Python wheel. Actively maintained with a release 7 days ago. Supports Python 3.10–3.13. Brings 15 runtime dependencies including heavy libraries (opencv-python, pandas, Pillow); optional extras available for video, document, and OpenAI integration if you don't need the full stack.

License in practice

Apache-2.0 permissive license allows commercial and private use without restriction, making it suitable for most production and research projects.

Quickstart

pip install vlmrun

from vlmrun.client import VLMRun
from vlmrun.common.utils import remote_image

client = VLMRun(api_key="<your-api-key>")
image = remote_image("https://example.com/image.jpg")
response = client.image.generate(images=[image], domain="document.invoice")
print(response)

Requires a valid VLM Run API key from app.vlm.run to authenticate requests.

Verify before relying

• Whether the 15 runtime dependencies are all required by default or if minimal installation is possible without optional extras.
• Performance characteristics and rate limits of the VLM Run API platform itself.
• Whether async support is available for all endpoints or only the agent completions endpoint.

Package facts

Evidence: vlmrun-0.7.3-py3-none-any.whl

Tags

More Libraries packages