buildfunctions 0.2.1

The Buildfunctions SDK for Agents: Hardware-isolated CPU and GPU Sandboxes for untrusted AI actions

The Buildfunctions SDK for Agents

 

Hardware-isolated execution environments for AI agents

Installation

pip install buildfunctions

Quick Start

1. Create an API Token

Get your API token at buildfunctions.com/settings

2. CPU Function

import asyncio
import os
from buildfunctions import Buildfunctions, CPUFunction

async def main():
    client = await Buildfunctions({"apiToken": os.environ["BUILDFUNCTIONS_API_TOKEN"]})

    deployed_function = await CPUFunction.create({
        "name": "my-cpu-function",
        "code": "./cpu_function_code.py",
        "language": "python",
        "memory": 128,
        "timeout": 30,
    })

    print(f"Endpoint: {deployed_function.endpoint}")

    await deployed_function.delete()

asyncio.run(main())

3. CPU Sandbox

import asyncio
import os
from buildfunctions import Buildfunctions, CPUSandbox

async def main():
    client = await Buildfunctions({"apiToken": os.environ["BUILDFUNCTIONS_API_TOKEN"]})

    sandbox = await CPUSandbox.create({
        "name": "my-cpu-sandbox",
        "language": "python",
        "code": "/path/to/code/cpu_sandbox_code.py",
        "memory": 128,
        "timeout": 30,
    })

    result = await sandbox.run()
    print(f"Result: {result}")

    await sandbox.delete()

asyncio.run(main())

4. GPU Function

import asyncio
import os
from buildfunctions import Buildfunctions, GPUFunction

async def main():
    client = await Buildfunctions({"apiToken": os.environ["BUILDFUNCTIONS_API_TOKEN"]})

    deployed_function = await GPUFunction.create({
        "name": "my-gpu-function",
        "code": "/path/to/code/gpu_function_code.py",
        "language": "python",
        "gpu": "T4",
        "vcpus": 30,
        "memory": "50000MB",
        "timeout": 300,
        "requirements": ["transformers==4.47.1", "torch", "accelerate"],
    })

    print(f"Endpoint: {deployed_function.endpoint}")

    await deployed_function.delete()

asyncio.run(main())

5. GPU Sandbox with Local Model

import asyncio
import os
from buildfunctions import Buildfunctions, GPUSandbox

async def main():
    client = await Buildfunctions({"apiToken": os.environ["BUILDFUNCTIONS_API_TOKEN"]})

    sandbox = await GPUSandbox.create({
        "name": "my-gpu-sandbox",
        "language": "python",
        "memory": 10000,
        "timeout": 300,
        "vcpus": 6,
        "code": "./gpu_sandbox_code.py",
        "model": "/path/to/models/Qwen/Qwen3-8B",
        "requirements": "torch",
    })

    result = await sandbox.run()
    print(f"Response: {result}")

    await sandbox.delete()

asyncio.run(main())

The SDK is currently in beta.