Core Concepts

Streaming Responses

Streaming allows you to receive tokens as soon as they are emitted by Cortiqa LPU inference nodes. Instead of waiting for full generation, chunks stream incrementally over Server-Sent Events (SSE) with sub-second time-to-first-token.

Overview

When streaming is enabled (stream: true), Cortiqa returns an event-stream containing incremental text delta chunks. All official Cortiqa SDKs feature built-in async iterators to handle streaming automatically.

Ultra-Fast Token Generation

Cortiqa LPU clusters generate reasoning tokens at high speed for openai/gpt-oss-120b. Streaming delivers the fastest interactive experience for your users.

Python SDK

Using the official cortiqa Python library:

stream_demo.py

1import cortiqa
2
3client = cortiqa.Client()
4
5stream = client.chat.create(
6    model="openai/gpt-oss-120b",
7    messages=[
8        {"role": "system", "content": "You are a helpful AI assistant powered by Cortiqa."},
9        {"role": "user", "content": "Explain quantum entanglement in simple terms."}
10    ],
11    stream=True
12)
13
14for chunk in stream:
15    if chunk.choices and chunk.choices[0].delta.content:
16        print(chunk.choices[0].delta.content, end="", flush=True)

TypeScript & Node.js SDK

Using the official @cortiqa/sdk:

stream.ts

1import { Cortiqa } from "@cortiqa/sdk";
2
3const cortiqa = new Cortiqa();
4
5async function main() {
6  const stream = await cortiqa.chat.createStream({
7    model: "openai/gpt-oss-120b",
8    messages: [
9      { role: "user", content: "Write a short poem about distributed computing." }
10    ],
11  });
12
13  for await (const chunk of stream) {
14    process.stdout.write(chunk.content || "");
15  }
16}
17
18main();

Go SDK

stream.go

1package main
2
3import (
4	"context"
5	"fmt"
6	"os"
7
8	"github.com/cortiqa-ai/cortiqa-go"
9)
10
11func main() {
12	client := cortiqa.NewClient(cortiqa.WithAPIKey(os.Getenv("CORTIQA_API_KEY")))
13
14	stream, err := client.Chat.CreateStream(context.Background(), &cortiqa.ChatRequest{
15		Model: "openai/gpt-oss-120b",
16		Messages: []cortiqa.Message{
17			{Role: "user", Content: "Give me 3 tips for cloud security."},
18		},
19	})
20	if err != nil {
21		panic(err)
22	}
23
24	for chunk := range stream {
25		fmt.Print(chunk.Delta)
26	}
27}

Raw HTTP / SSE

Direct HTTP request sending "stream": true to https://api.cortiqa.co/api/v1/chat/completions:

Terminal

curl https://api.cortiqa.co/api/v1/chat/completions \
  -H "Authorization: Bearer sk-cortiqa-YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-oss-120b",
    "messages": [{"role": "user", "content": "Hi"}],
    "stream": true
  }'

Raw response chunks arrive as Server-Sent Events:

Server-Sent Events

data: {"id":"chatcmpl-91a...","object":"chat.completion.chunk","model":"openai/gpt-oss-120b","choices":[{"index":0,"delta":{"role":"assistant","content":""},"finish_reason":null}]}

data: {"id":"chatcmpl-91a...","object":"chat.completion.chunk","model":"openai/gpt-oss-120b","choices":[{"index":0,"delta":{"content":"Hello"},"finish_reason":null}]}

data: {"id":"chatcmpl-91a...","object":"chat.completion.chunk","model":"openai/gpt-oss-120b","choices":[{"index":0,"delta":{"content":"! How"},"finish_reason":null}]}

data: {"id":"chatcmpl-91a...","object":"chat.completion.chunk","model":"openai/gpt-oss-120b","choices":[{"index":0,"delta":{"content":" can I help you today?"},"finish_reason":"stop"}]}

data: [DONE]

Was this page helpful?

Messages API

Context Windows