API documentation

1. Base URL and authentication

Create an API key in the dashboard, then send it as a bearer token:

Base URL:       https://api.bikai.bikiran.win/v1
Authorization:  Bearer sk-bik-...

2. Chat completions

curl https://api.bikai.bikiran.win/v1/chat/completions \
  -H "Authorization: Bearer $BIKAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-4o-mini",
    "messages": [{"role": "user", "content": "Write a haiku about Dhaka."}]
  }'

import OpenAI from "openai";

const client = new OpenAI({ baseURL: "https://api.bikai.bikiran.win/v1", apiKey: process.env.BIKAI_API_KEY });

const stream = await client.chat.completions.create({
  model: "anthropic/claude-3.5-sonnet",
  messages: [{ role: "user", content: "Explain streaming in one sentence." }],
  stream: true,
});
for await (const chunk of stream) process.stdout.write(chunk.choices[0]?.delta?.content ?? "");

3. Endpoints

• GET /models — models available to your key, with per-token prices in AI Credit.
• GET /models/{id} — a single model.
• POST /chat/completions — chat, streaming (SSE), tools / function calling, JSON mode, vision.
• POST /completions — legacy text completions.

Model ids look like openai/gpt-4o-mini. Set stream_options.include_usage to receive a final usage chunk when streaming. n greater than 1 is not supported.

4. Billing

Each request is charged from your AI Credit for its actual usage on the provider that served it. Model prices on the Models page and in GET /models are approximate, and actual charges are usually at or below them. The exact charge for every request is shown in Usage. Failed requests are not charged; if you disconnect during a stream you are charged only for tokens already generated. Every request appears in Usage.

5. Errors

Errors use the OpenAI format: {"error": {"message", "type", "param", "code"}}.