Dolphin api Quickstart

Base URL and Authentication

Your API key authenticates every request to our endpoint. Use the base URL HTTPS://api.apidolphin.com/v1 with any OpenAI-compatible SDK. The model identifier is "uncensored" for all completions. This setup ensures your code works with our infrastructure without modification.

We support standard parameters like temperature and top_p. You can enable JSON mode or function calling as needed. Errors and refusals do not consume your prepaid credit. Only actual token usage is charged at our transparent rates.

Send Your First Request

Send a basic chat completion to verify your connection. Use the curl command below to test the endpoint. Replace YOUR_API_KEY with your actual key from the dashboard. The response returns the model's text output.

This request uses the standard OpenAI format. You can adjust system prompts or user messages as required. The API returns a response object containing the generated text. Verify the output matches your expectations before scaling up.

Example usage:

Copycurl https://api.apidolphin.com/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "uncensored",
    "messages": [{"role": "user", "content": "Convert the attached JSON payload to YAML format"}]
  }'

Python SDK Integration

Use the official OpenAI Python library to interact with our API. Set the base URL and API key in your client configuration. The code snippet below demonstrates a simple text generation call.

Install the library via pip. Then initialize the client with your credentials. Call chat.completions.create with the model name "uncensored". This approach keeps your code portable across different LLM providers.

Code example:

Copyfrom openai import OpenAI

client = OpenAI(base_url="https://api.apidolphin.com/v1", api_key="YOUR_KEY")

resp = client.chat.completions.create(
    model="uncensored",
    messages=[{"role": "user", "content": "Summarise this thread without softening it."}],
)
print(resp.choices[0].message.content)

Node SDK Integration

The Node SDK follows the same configuration pattern. Initialize the OpenAI client with your base URL and key. Send a completion request using the "uncensored" model. This method is ideal for server-side applications.

Ensure you handle the response data correctly. The API returns a standard completion object. You can access the generated text from the response object. This setup works with Express, Fastify, or any Node.js framework.

Implementation example:

Copyimport OpenAI from "openai";

const client = new OpenAI({ baseURL: "https://api.apidolphin.com/v1", apiKey: process.env.API_KEY });

const resp = await client.chat.completions.create({
  model: "uncensored",
  messages: [{ role: "user", content: "Draft a villain monologue for my game." }],
});
console.log(resp.choices[0].message.content);

Handle Streaming Responses

Enable streaming for real-time token delivery. Set the stream parameter to true in your request. The API returns data chunks as they are generated. This reduces perceived latency for end-users.

Process each chunk in your application logic. The last chunk contains the final token usage details. Streaming works with all supported parameters. It is ideal for chat interfaces and live text generation.

Streaming example:

Copystream = client.chat.completions.create(
    model="uncensored",
    messages=[{"role": "user", "content": "Tell the story in second person."}],
    stream=True,
)
for chunk in stream:
    if chunk.choices and chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="", flush=True)

Limits, Errors, and Context Window

Your account has a 100,000 token context window. This includes both prompt and completion tokens. The maximum output per request is 16,000 tokens. If you do not set max_tokens, the limit is 2,048 tokens.

Rate limits allow 300 requests per minute per key. You can have 8 concurrent requests. The request body size limit is 8 MB. Handle HTTP 401 for invalid keys, 402 for insufficient credit, and 429 for rate limits. Errors are free, so you can retry without cost.