import { streamText } from "ai";
import { createOpenAICompatible } from "@ai-sdk/openai-compatible";

const provider = createOpenAICompatible({
  name: "llama.cpp",
  apiKey: "local",
  baseURL: "http://127.0.0.1:8080/v1",
  includeUsage: true, // Include usage information in streaming responses
});

const { fullStream, reasoningText, text } = streamText({
  model: provider("qwen3.5-35b-a3b"),
  prompt: Bun.argv[2]!,
});

for await (const part of fullStream) {
  if (part.type === "reasoning-start") {
    process.stdout.write("\u001b[2m\n    ");
  } else if (part.type === "reasoning-delta") {
    process.stdout.write(part.text.replaceAll('\n', '\n    '));
  } else if (part.type === "reasoning-end") {
    process.stdout.write("\n\u001b[22m");
  } else if (part.type === "text-start") {
    process.stdout.write("\n");
  } else if (part.type === "text-delta") {
    process.stdout.write(part.text);
  } else if (part.type === "text-end") {
    process.stdout.write("\n");
  }
}

const reasoning = await reasoningText;
const response = await text;

Bun.file("reasoning.md").write(reasoning ?? "");
Bun.file("response.md").write(response);