Node / TypeScript SDK

For Node.js and TypeScript projects, two official packages are recommended:

• OpenAI compatible: openai (official)
• Anthropic native: @anthropic-ai/sdk (official)

Examples are based on the official READMEs pulled via Context7. Refer to the upstream repos for the latest API: openai/openai-node, anthropics/anthropic-sdk-typescript.

Setup

export SAKRYLLE_API_KEY="sk-xxxxxxxxxxxxxxxx"

Node.js 18+ is required (built-in fetch). Both SDKs are written in TypeScript but work fine in plain JavaScript projects.

OpenAI compatible

Install

npm install openai
# or pnpm add openai / yarn add openai

Minimal usage

import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: process.env.SAKRYLLE_API_KEY,
  baseURL: 'https://api.sakrylle.com/v1',
});

const completion = await client.chat.completions.create({
  model: 'gpt-5.6-sol',
  messages: [
    { role: 'system', content: 'You are a concise assistant.' },
    { role: 'user', content: 'Explain RAG in one sentence.' },
  ],
});

console.log(completion.choices[0].message.content);

OPENAI_API_KEY + OPENAI_BASE_URL env vars also work — new OpenAI() reads them automatically.

Streaming

import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: process.env.SAKRYLLE_API_KEY,
  baseURL: 'https://api.sakrylle.com/v1',
});

const stream = await client.chat.completions.create({
  model: 'gpt-5.6-sol',
  messages: [{ role: 'user', content: 'Count to 5.' }],
  stream: true,
});

for await (const chunk of stream) {
  process.stdout.write(chunk.choices[0]?.delta?.content ?? '');
}

Tool calling

import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: process.env.SAKRYLLE_API_KEY,
  baseURL: 'https://api.sakrylle.com/v1',
});

const tools = [
  {
    type: 'function' as const,
    function: {
      name: 'get_weather',
      description: 'Look up the current weather for a city',
      parameters: {
        type: 'object',
        properties: {
          city: { type: 'string', description: 'City name' },
        },
        required: ['city'],
      },
    },
  },
];

const resp = await client.chat.completions.create({
  model: 'gpt-5.6-sol',
  messages: [{ role: 'user', content: "What's the weather in Beijing today?" }],
  tools,
});

const choice = resp.choices[0];
if (choice.message.tool_calls?.length) {
  for (const call of choice.message.tool_calls) {
    if (call.type === 'function') {
      console.log('call:', call.function.name, JSON.parse(call.function.arguments));
    }
  }
} else {
  console.log(choice.message.content);
}

The official SDK also offers client.chat.completions.runTools(...) for multi-turn tool loops — see openai-node helpers.md.

Anthropic native

Install

npm install @anthropic-ai/sdk

Minimal usage

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

const client = new Anthropic({
  apiKey: process.env.SAKRYLLE_API_KEY,
  baseURL: 'https://api.sakrylle.com',
});

const message = await client.messages.create({
  model: 'claude-haiku-4-5-20251001',
  max_tokens: 1024,
  messages: [{ role: 'user', content: 'Introduce Claude in one sentence.' }],
});

console.log(message.content);

Use https://api.sakrylle.com as baseURL; the SDK appends /v1/messages automatically. ANTHROPIC_API_KEY / ANTHROPIC_BASE_URL env vars also work.

Streaming

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

const client = new Anthropic({
  apiKey: process.env.SAKRYLLE_API_KEY,
  baseURL: 'https://api.sakrylle.com',
});

const stream = client.messages.stream({
  model: 'claude-haiku-4-5-20251001',
  max_tokens: 1024,
  messages: [{ role: 'user', content: 'Write a four-line poem.' }],
});

for await (const event of stream) {
  if (event.type === 'content_block_delta' && event.delta.type === 'text_delta') {
    process.stdout.write(event.delta.text);
  }
}

const finalMessage = await stream.finalMessage();
console.log('\n[done]', finalMessage.usage);

Error handling

import OpenAI, { APIError, AuthenticationError, RateLimitError } from 'openai';

const client = new OpenAI({
  apiKey: 'bad-key',
  baseURL: 'https://api.sakrylle.com/v1',
});

try {
  await client.chat.completions.create({
    model: 'gpt-5.6-sol',
    messages: [{ role: 'user', content: 'hi' }],
  });
} catch (err) {
  if (err instanceof AuthenticationError) {
    console.error('Key invalid or revoked');
  } else if (err instanceof RateLimitError) {
    console.error('Rate limited, retry later');
  } else if (err instanceof APIError) {
    console.error('Upstream error:', err.status, err.message);
  } else {
    throw err;
  }
}

For a full error table, see Errors.