Help Peer Python SDK

A Python SDK for asynchronously sharing large file dumps (AI model weights) via the Help Peer network.

Installation

pip install helppeer

Quick Start

import helppeer

# Send a model directory
code = helppeer.send("./my-model", name="Llama-3-70B")
print(f"Transfer code: {code}")
# Share this code with the recipient. Data lives for 24 hours.

# Receive a transfer
helppeer.receive("38-vortex-xenon", output_dir="./received")

Configuration

import helppeer

# Use custom relay and storage nodes
helppeer.configure(
    relay="https://relay.helppeer.dev",
    storage_nodes=["https://node1.helppeer.dev", "https://node2.helppeer.dev"],
)

# Or use environment variables:
# HELPEER_RELAY_URL, HELPEER_STORAGE_NODES (comma-separated)

Agent Integration

import helppeer
import json

# Send and get structured result
result = helppeer.send("./model-weights", name="my-model", return_details=True)
# result = {"code": "38-vortex-xenon", "transfer_name": "my-model", "files": 4, "total_bytes": 1056}

# Receive with progress callback
def on_progress(file_path, segment, total_segments):
    print(f"  {file_path}: {segment}/{total_segments} segments")

helppeer.receive("38-vortex-xenon", output_dir="./received", progress=on_progress)

CLI

The SDK also installs a helppeer CLI:

helppeer send ./my-model --name "Llama-3-70B"
helppeer receive 38-vortex-xenon --output ./received