Nataris — built a P2P inference network where Android phones serve API requests

[Sharrmavishal]

Built this over three months with a two-person team. Since text-generation-webui supports OpenAI-compatible API backends, figured this might be relevant here.

What Nataris is

A P2P inference marketplace. Android phones run open-weight models locally (Qwen 2.5 0.5B, Llama 3.2 1B) and serve API requests via a standard OpenAI-compatible endpoint. Phone owners get paid per token. You get inference without managing any servers.

No prompt logging. No content filtering. No model training on your queries.

Using with text-generation-webui

In the OpenAI API extension settings, point to:

API base: https://api.nataris.ai/v1
API key: your Nataris API key

Model aliases:

• nataris-fast — Qwen 2.5 0.5B (~5s)
• nataris-balanced — Llama 3.2 1B (~15-20s)

Keep streaming enabled — device cold-starts can cause timeouts on non-streaming requests.

Where we are

21 provider devices on the network, 2,775 inference jobs completed, 350K+ tokens processed in closed beta. Android app just went live on Google Play.

Works well for anything where 5–20s latency is acceptable.

$5 free credits on signup, no card needed.

API: https://api.nataris.ai/v1
Docs: https://api.nataris.ai/docs
Provider app (earn by running models on your Android): https://play.google.com/store/apps/details?id=ai.nataris.app