Ollama API: Switch your client in three lines

Prerequisites and Base URL

To use this uncensored LLM API, you need an account with an API key and an OpenAI-compatible SDK installed in your environment. The base URL for all requests is https://api.ollamaapi.com/v1. This endpoint is designed as a drop-in replacement for existing OpenAI clients, meaning you only need to update the base URL and API key in your configuration. The API supports text-in, text-out completions with streaming via Server-Sent Events (SSE) and tool calling. It does not support embeddings, image generation, or fine-tuning. Ensure your SDK version supports these features for the best experience. You will use the model ID uncensored for all completions.

Authentication and First Request

Authentication is handled via a Bearer token in the Authorization header. Sign up on the site to receive your key immediately; no credit card is required for the trial. Use the following cURL command to send your first message. Replace YOUR_API_KEY with your actual key. The request targets the standard chat completions endpoint. If successful, you will receive a text response from the uncensored model. Ensure you set the model field to uncensored to utilize the specific large language model hosted on our GPU servers.

Copycurl https://api.ollamaapi.com/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "uncensored",
    "messages": [{"role": "user", "content": "Write a blunt product review of a cheap VPN."}]
  }'

Remember that prompts are not used for training your data. The API enforces a hard limit on sexual content involving minors, but otherwise allows lawful adult and controversial topics without refusal.

Python SDK Integration

Using the official OpenAI Python SDK is the most straightforward way to integrate. Install the library via pip, then initialize the client with the custom base URL and your API key. This setup allows you to use the same client code you might use for GPT-4, but directed at our uncensored model. The code below demonstrates a simple chat completion request. You can adjust the max_tokens and temperature parameters as needed. The response object will contain the generated text in response.choices[0].message.content.

Copyfrom openai import OpenAI

client = OpenAI(base_url="https://api.ollamaapi.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)

This approach works because our API is fully OpenAI-compatible. You do not need a separate client library specifically for our service.

Node.js SDK Setup

For JavaScript or TypeScript projects, the OpenAI Node SDK works seamlessly with our endpoint. Install the package and configure the client with the baseURL pointing to our domain. Pass your API key in the apiKey field. The request structure mirrors the OpenAI format exactly. This allows you to drop our API into existing Node.js applications without significant refactoring. The model ID remains uncensored. You can manage your API key via the dashboard if you need to regenerate it.

Copyimport OpenAI from "openai";

const client = new OpenAI({ baseURL: "https://api.ollamaapi.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);

Ensure your environment variables are set correctly to avoid exposing your key in source control. The API key is shown immediately upon signup.

Streaming Responses

For applications requiring real-time token generation, the API supports Server-Sent Events (SSE). Set the stream parameter to true in your request. The server will send a series of chunks, each containing a partial delta of the response. This is ideal for chat interfaces where you want to display text as it is generated. The streaming behavior is identical to the OpenAI streaming API, so existing streaming logic should work with minimal changes. Handle the stream until the final chunk indicates the response is complete.

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)

Streaming does not affect the token pricing or the context window limits.

Limits, Errors, and Context

The API has a 100,000 token context window, combining prompt and completion. This allows for extensive document processing or long conversations. Rate limits are set to 300 requests per minute per key. If you exceed this, you will receive a 429 error. Authentication errors return a 401 status, usually due to an invalid or revoked key. Insufficient funds result in a 402 error. Request bodies are limited to 8 MB. There is no subscription; you pay as you go with prepaid credits that never expire. You can top up starting at $20. Credits do not expire, so you can pause usage indefinitely without losing your balance.