buildfunctions 0.2.2

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

from buildfunctions import Buildfunctions, CPUFunction

client = await Buildfunctions({"apiToken": 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()

3. CPU Sandbox

from buildfunctions import Buildfunctions, CPUSandbox

client = await Buildfunctions({"apiToken": 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()

4. GPU Function

from buildfunctions import Buildfunctions, GPUFunction

client = await Buildfunctions({"apiToken": 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()

5. GPU Sandbox with Local Model

from buildfunctions import Buildfunctions, GPUSandbox

client = await Buildfunctions({"apiToken": 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()

The SDK is currently in beta.