API Reference

Complete documentation for the Furcate decentralized federated learning platform

Base URL

https://api.furcate.io

All API requests should be made to this base URL with the appropriate endpoint path

System Architecture

Furcate is a decentralized federated learning platform that enables collaborative AI training across edge devices without centralizing sensitive data. The system integrates multiple technologies:

Blockchain Layer

Minima blockchain for trustless coordination and proof of training

P2P Messaging

Maxima protocol for secure node-to-node communication

Cloud Storage

GCP buckets via Tenzro Network for gradient and model artifact storage

Secure Compute

TEE attestation for verifiable computation integrity

Core Concepts

Fleets

A fleet represents a collaborative learning group. Each fleet has an owner, privacy parameters, and defines the model architecture for federated training.

Training Rounds

Rounds coordinate synchronized training cycles. Each round collects contributions from nodes, aggregates updates using FedAvg or other methods, and produces a new global model.

Contributions

Individual model updates from edge devices. Each contribution includes gradient updates, privacy metrics, and trust attestations stored in distributed storage.

Aggregation

The process of combining verified contributions into an improved global model using weighted averaging and privacy-preserving techniques.

Autonomous Coordinator

Background service that automatically monitors rounds, triggers aggregation when ready, distributes blockchain rewards, and creates next training rounds without manual intervention.

Protocol Flow

  1. 01

    Fleet Creation

    Initialize a new federated learning fleet with privacy parameters and model architecture

  2. 02

    Round Initialization

    Start a training round with target participants and aggregation parameters

  3. 03

    Local Training

    Edge devices train on local data and compute gradient updates

  4. 04

    Contribution Submission

    Nodes submit encrypted gradients with privacy proofs and trust attestations

  5. 05

    Autonomous Aggregation

    Coordinator automatically detects round readiness and triggers FedAvg aggregation

  6. 06

    Reward Distribution

    Automatic blockchain payments to contributors based on trust level and sample count

  7. 07

    Next Round Creation

    Coordinator automatically initializes the next training round with updated model

Rate Limits

Standard Tier

100 requests per minute

Suitable for development and small-scale deployments

Enterprise Tier

10,000 requests per minute

For production deployments with multiple fleets