Calling LLMs from Code: The API Basics

{
  id: "msg_01XFDUDYJgAACzvnptvVoYEL",
  type: "message",
  role: "assistant",
  content: [{ type: "text", text: "4" }],
  model: "claude-3-5-sonnet-20241022",
  stop_reason: "end_turn",    // or "max_tokens" if you hit the ceiling
  usage: {
    input_tokens: 14,
    output_tokens: 2
  }
}

import Anthropic from "@anthropic-ai/sdk";

const client = new Anthropic();

const stream = await client.messages.stream({
  model: "claude-3-5-sonnet-20241022",
  max_tokens: 512,
  messages: [{ role: "user", content: "Explain photosynthesis in 3 sentences." }],
});

for await (const chunk of stream) {
  if (
    chunk.type === "content_block_delta" &&
    chunk.delta.type === "text_delta"
  ) {
    process.stdout.write(chunk.delta.text);  // arrives token-by-token
  }
}

const final = await stream.getFinalMessage();
console.log("\nTokens used:", final.usage.input_tokens + final.usage.output_tokens);

import anthropic

client = anthropic.Anthropic()

with client.messages.stream(
    model="claude-3-5-sonnet-20241022",
    max_tokens=512,
    messages=[{"role": "user", "content": "Explain photosynthesis in 3 sentences."}],
) as stream:
    for text in stream.text_stream:
        print(text, end="", flush=True)

final = stream.get_final_message()
print(f"\nTokens: {final.usage.input_tokens + final.usage.output_tokens}")

import Anthropic from "@anthropic-ai/sdk";
import * as readline from "node:readline/promises";

const client = new Anthropic();
const SYSTEM = "You are a concise assistant. Answer in 2–3 sentences unless asked for more.";
const history: Anthropic.MessageParam[] = [];

const rl = readline.createInterface({ input: process.stdin, output: process.stdout });
console.log("Chat started. Ctrl-C to quit.\n");

while (true) {
  const userInput = await rl.question("You: ");
  history.push({ role: "user", content: userInput });

  // Rough context trim — keep system prompt intact, drop oldest pairs
  while (history.reduce((n, m) => n + (m.content as string).length, 0) > 600_000) {
    history.splice(0, 2);
  }

  process.stdout.write("Assistant: ");
  const stream = await client.messages.stream({
    model: "claude-3-5-sonnet-20241022",
    max_tokens: 512,
    system: SYSTEM,
    messages: history,
  });

  let assistantReply = "";
  for await (const chunk of stream) {
    if (chunk.type === "content_block_delta" && chunk.delta.type === "text_delta") {
      process.stdout.write(chunk.delta.text);
      assistantReply += chunk.delta.text;
    }
  }

  process.stdout.write("\n\n");
  history.push({ role: "assistant", content: assistantReply });
}