> ## Documentation Index
> Fetch the complete documentation index at: https://docs.kymaapi.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Build a Chatbot

> Streaming multi-turn chatbot with conversation history.

## Best Model for This

| Model              | Why                                 | Cost per 1K messages |
| ------------------ | ----------------------------------- | -------------------- |
| `qwen-3.6-plus`    | Best quality, handles nuance well   | \~\$0.30             |
| `llama-3.3-70b`    | Fastest response, great all-rounder | \~\$0.80             |
| `gemini-2.5-flash` | Cheapest at scale, 1M context       | \~\$0.20             |

Costs assume \~300 tokens input + \~200 tokens output per message exchange.

## Quick Start

<CodeGroup>
  ```python Python theme={null}
  from openai import OpenAI

  client = OpenAI(
      base_url="https://kymaapi.com/v1",
      api_key="ky-your-api-key"
  )

  messages = [{"role": "system", "content": "You are a helpful assistant."}]

  def chat(user_input: str) -> str:
      messages.append({"role": "user", "content": user_input})
      stream = client.chat.completions.create(
          model="qwen-3.6-plus",
          messages=messages,
          stream=True
      )
      response = ""
      for chunk in stream:
          content = chunk.choices[0].delta.content
          if content:
              print(content, end="", flush=True)
              response += content
      print()
      messages.append({"role": "assistant", "content": response})
      return response

  chat("What is machine learning?")
  chat("Can you give me a simple example?")
  ```

  ```javascript JavaScript theme={null}
  import OpenAI from "openai";

  const client = new OpenAI({
    baseURL: "https://kymaapi.com/v1",
    apiKey: "ky-your-api-key",
  });

  const messages = [{ role: "system", content: "You are a helpful assistant." }];

  async function chat(userInput) {
    messages.push({ role: "user", content: userInput });
    const stream = await client.chat.completions.create({
      model: "qwen-3.6-plus",
      messages,
      stream: true,
    });
    let response = "";
    for await (const chunk of stream) {
      const content = chunk.choices[0]?.delta?.content;
      if (content) {
        process.stdout.write(content);
        response += content;
      }
    }
    console.log();
    messages.push({ role: "assistant", content: response });
    return response;
  }

  await chat("What is machine learning?");
  await chat("Can you give me a simple example?");
  ```
</CodeGroup>

## Tips & Best Practices

* **Stream always** — users perceive streamed responses as 3-5x faster even at the same token speed.
* **Cap history length** — trim `messages` to the last 10-20 turns or \~4K tokens to keep latency low and costs predictable.
* **System prompt sets personality** — define tone, scope, and any constraints in the first system message.
* **Use `qwen-3.6-plus` for quality, `llama-3.3-70b` for speed** — swap the model string; the code is identical.

## Cost Estimate

| Volume           | Model              | Monthly cost |
| ---------------- | ------------------ | ------------ |
| 1K messages/day  | `qwen-3.6-plus`    | \~\$9/month  |
| 1K messages/day  | `llama-3.3-70b`    | \~\$24/month |
| 10K messages/day | `gemini-2.5-flash` | \~\$60/month |

Assumes 300 tokens input + 200 tokens output per exchange. Longer conversations cost more.

## Next Steps

* [Streaming](/guides/streaming) — deeper streaming patterns
* [Model Aliases](/guides/model-aliases) — use `best`, `fast`, `balanced` shortcuts
* [Prompt Caching](/guides/prompt-caching) — cache system prompts to cut costs by up to 90%