Loading integrations

import OpenAI from "openai" const client = new OpenAI({ apiKey: process.env.TSUBASA_API_KEY, baseURL: "https://api.tsubasa.sh/v1", }) const stream = await client.chat.completions.create({ model: "tsubasa-pro", messages: [ { role: "user", content: "Review this repository and propose a patch." }, ], stream: true, })

SDK INTEGRATION

Verified

OpenAI JavaScript SDK

Use Tsubasa with the official JavaScript SDK.

Covered by a reproducible compatibility test in this repository.

API dialect: OpenAI Chat Completions
Tested version: openai 7.4.0
Streaming: Verified with server-sent events
Verified on: 2026-08-13

Setup

Uses an example ID from the current environment. It is not a recommendation unless the capability panel says so.

import OpenAI from "openai"

const client = new OpenAI({
  apiKey: process.env.TSUBASA_API_KEY,
  baseURL: "https://api.tsubasa.sh/v1",
})

const stream = await client.chat.completions.create({
  model: "tsubasa-pro",
  messages: [
    { role: "user", content: "Review this repository and propose a patch." },
  ],
  stream: true,
})

Current limitations

Only GET /v1/models and POST /v1/chat/completions are implemented.
Responses, Assistants, Files, Images, Audio, Embeddings, and Batch endpoints are not available.
Tool calls and structured outputs are experimental until live model/provider qualification passes.