actava

ActAVA SDK

Enterprise AI orchestration platform for healthcare organizations. Build, deploy, and manage AI agents on ActAVA's managed cloud platform.

Installation

pip install actava

Quick Start

import actava as av
from actava.langchain_openai import ChatOpenAI
from actava.langgraph import StateGraph, END

# Initialize ActAVA with your project credentials
av.init(project="your-project-id", api_key="your-api-key")

# Use LangChain/LangGraph as usual
llm = ChatOpenAI(model="gpt-4o")

def my_function(state):
    return {"result": "processed", **state}

# Create and compile graph
graph = StateGraph(dict)
graph.add_node("process", my_function)
graph.add_edge("process", END)
compiled = graph.compile()

# Export graph for deployment
from actava.graph import export
spec = export(graph)
print(spec.model_dump_json(indent=2))

Features

• Zero API Changes: Import actava.langchain and actava.langgraph with no code changes
• Auto Telemetry: Automatic instrumentation with configurable sampling
• Graph Export: Export LangGraph DAGs to JSON GraphSpec for deployment
• Runtime Service: FastAPI-based runtime service for local testing and cloud deployment
• Healthcare Focus: Built with healthcare compliance and security in mind
• Managed Cloud: Deploy to ActAVA's managed cloud platform with actava-cli

Deployment

Use the ActAVA CLI to deploy your agents to our managed cloud:

# Install CLI
pip install actava-cli

# Initialize project
actava init

# Deploy agent
actava push --manifest agent.manifest.yaml

Documentation

For full documentation, visit docs.actava.ai

Support

• Commercial Support: support.actava.ai
• Documentation: docs.actava.ai
• Enterprise Sales: actava.ai

License

Proprietary software. Licensed for use by ActAVA customers only.