Chat completion (openai)

post

Creates a completion for chat messages

Body

modelstring · enumOptionalDefault: kirha:cryptoPossible values:

streambooleanOptionalDefault: false

temperaturenumber · max: 2Optional

top_pnumber · max: 1Optional

Responses

200

An openai compatible chat completion response (JSON or SSE stream)

application/jsontext/event-stream

post

HTTPcURLJavaScriptPython

POST /chat/v1/openai/chat/completions HTTP/1.1
Host: api.kirha.ai
Content-Type: application/json
Accept: */*
Content-Length: 113

{
  "model": "kirha:crypto",
  "messages": [
    {
      "role": "system",
      "content": "text"
    }
  ],
  "stream": false,
  "temperature": 1,
  "top_p": 1
}

200

An openai compatible chat completion response (JSON or SSE stream)

{
  "id": "text",
  "object": "text",
  "created": 1,
  "choices": [
    {
      "index": 1,
      "message": {
        "role": "system",
        "content": "text"
      },
      "finish_reason": "text"
    }
  ]
}

OpenAi SDK

You can use the completion endpoint with the official OpenAI sdk:

import OpenAI from "openai";

const openaiClient = new OpenAI({
  apiKey: "<kirha_api_key>",
  baseURL: "https://api.kirha.ai/chat/v1/openai",
});

const response = await openaiClient.chat.completions.create({
  model: "kirha-flash:crypto",
  messages: [
    { role: "system", content: "You are a helpful crypto assistant." },
    { role: "user", content: "give me the balance of vitalik.eth" }
  ],
  stream: true,
});

for await (const chunk of response) {
  if (chunk.choices[0].delta.content) {
    console.log(chunk.choices[0].delta.content);
  }
}