Targon Chat Completions Streaming (HTTP + SSE)
Version 1.0.0
AsyncAPI 2.6 description of Targon's **chat completion streaming** surface. Targon is a decentralized AI inference platform operated as Bittensor Subnet 4 by Manifold Labs. It does not publish a WebSocket API. The only asynchronous / event-style transport documented for its OpenAI-compatible API at https://api.targon.com/v1 is **HTTP Server-Sent Events (SSE)** delivered over the same REST endpoint (`POST /chat/completions`) when the request body sets `stream: true`. SSE is a one-way, server-to-client HTTP streaming channel; it is **not** WebSocket. Following the OpenAI-compatible convention Targon implements, tokens are sent as data-only server-sent events as they become available, with the stream terminated by a `data: [DONE]` message. This AsyncAPI document models only the streamed events emitted by Targon's SSE response. The request body itself (model, messages, etc.) is modeled in the companion OpenAPI document at `openapi/targon-openapi.yml`.
Channels
Messages
Servers
api.targon.com/v1