Federated Learning: Train AI Across Browsers, Phones and GPUs

Guide6 min read · Updated 2026-06-01

How federated learning lets every device — browser, phone, CPU, GPU — help train the same model. What works, what does not, and how a universal small-model pool is built.

What federated learning is

Federated learning trains a shared model across many devices without moving the model to one place. Each device trains on its data locally and sends only weight updates, which are averaged into a global model. Google uses it on phones (Gboard); frameworks like Flower and FedML generalize it.

The honest limit: the model must fit the device

A device can only train a model it can hold and run. Small models (compact CNNs, MLPs, classifiers) fit in a browser via WebGPU/TensorFlow.js, on phones, CPUs and GPUs — so every device can train the same small model.

Large models (YOLO, ResNet-50, 7B+ LLMs) cannot run in a browser. They need real GPUs. No trick removes that memory limit — so big-model training stays on the GPU tier while light devices help with smaller models and data tasks.

How "every device trains" works in practice

For a small model, the server initializes shared weights, splits the data, and any device — browser included — trains its slice in its native runtime (TensorFlow.js in the browser, PyTorch on a GPU) and uploads weights as JSON. The server averages them in pure JavaScript (FedAvg) and runs more rounds.

Because weights are tiny numeric arrays, browsers and GPUs interoperate, and contributors are paid by verified contribution.

Project Huginn's universal track

Project Huginn includes a universal small-model federated track: open a browser worker page, paste your token, and your device joins the pool and trains. Capability-aware routing keeps heavy jobs on GPUs and light jobs on any device — putting the maximum number of devices to work.

Frequently asked questions

Can a browser really train an AI model?
Yes, small models — via WebGPU/TensorFlow.js. Browsers cannot train large models like YOLO or LLMs; those need real GPUs.
Is my data shared with other devices?
In federated learning only weight updates are exchanged, not raw data; each device trains on its own slice.
Try it on Project Huginn

Train & fine-tune AI models on a distributed GPU pool — from €0.21/HU, cost estimated up front.

Start training →
More guides
What Is Distributed AI Training? A Plain-English Guide (2026)How to Train AI Models Cheaply Using Idle GPUsLoRA vs QLoRA: Which Fine-Tuning Method Should You Use?GPU Cloud Alternatives in 2026: Marketplaces & Distributed Pools Compared