Pure API

Dolphin uncensored API Integration Guide

Drop in our base_url to access an uncensored AI chat-completions endpoint with a 100,000-token context window for immediate integration.

OpenAI-compatible chat-completions endpoint
$0.15 per 1M input tokens
No content filtering for adults

Try with trial credit View documentation

Test with one call

curl https://api.dolphinuncensored.com/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "uncensored",
    "messages": [{"role": "user", "content": "fetch the list of installed python packages"}]
  }'

from openai import OpenAI

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

import OpenAI from "openai";

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

Core features

OpenAI Compatible

Works with existing OpenAI SDKs by changing only the base_url and API key.

Large Context

Supports 100,000 tokens total with 16,000 output tokens per request.

Streaming Support

Full streaming via SSE with token usage details in the final chunk.

JSON Mode

Enforce structured outputs using response_format json_object for reliable parsing.

Function Calling

Native support for tools and tool_choice parameters in your prompts.

Privacy First

Prompts are not used for training and no phone number is required.

Setup guide

Create Account

Add Credit

Top up with crypto to start using the API with bonus credits.

Integrate Code

Update your base_url and start sending requests to the endpoint.

Community projects

Content Generation

Generate creative text without refusals for lawful adult or fictional topics.

Data Processing

Process large documents using the extensive context window for analysis.

Dev Testing

Test function calling and JSON modes with reliable structured responses.

Research Models

Experiment with different models without vendor lock-in or extra fees.

Common mistakes people make in this niche and how to avoid them

Developers often confuse context limits with output caps, assuming a 100,000-token window allows infinite responses. It does not. Our API caps completion at 16,000 tokens by default. Another error is expecting embeddings or fine-tuning support. This is a pure chat-completions endpoint. Finally, many forget that uncensored means fewer refusals, not fewer hallucinations. Set your parameters explicitly.

dolphin uncensored: The Uncensored AI API for Developers

Our API serves a single, pure model that answers without standard refusals.

Content-Agnostic Reasoning

Get consistent outputs for controversial topics without manual prompt engineering.

Cost-Efficient Scaling

Pay only for tokens used, starting at $0.15 per million input tokens.

Predictable Limits

Enjoy a 100,000-token context window with no hidden subscription fees.

What developers actually ask about this topic

Is Dolphin the best uncensored AI option?

It is a leading uncensored model for text. Venice Mistral adds vision. Huihui-ai releases other uncensored variants. Dolphin offers a quick setup with Bionic. Our API strips multi-model routing bloat. It serves one model. You drop in a base_url. You access high-uncensored reasoning immediately without vendor lock-in. It is pure text input and output.

Moving from another OpenAI-compatible provider in three steps

Switching is trivial: update your base URL and API key. First, replace your existing endpoint with https://api.dolphinuncensored.com/v1. Second, paste your new key from the dashboard into your SDK configuration. Third, send a test request to /v1/models to verify connectivity. No code changes are needed for function calling or streaming. This approach works with any standard OpenAI-compatible client, ensuring immediate compatibility without modifying your application logic.

How this differs from a typical alternative on 4-6 concrete signs

We charge only for real tokens, not subscriptions. Input costs $0.15 per million tokens; output runs $0.50 per million. No hidden fees apply when errors occur. The prepaid credit never expires, ensuring long-term value without monthly renewals.

Who uses it, in which real situations; and plainly who it is NOT for

Developers use this uncensored AI API for creative writing, jailbreak testing, and research requiring fewer refusals. It fits projects needing raw model output without extra features.

Best for: pure text generation with function calling.
Not for: teams needing image or audio generation.
Not for: apps requiring embeddings or fine-tuning.
Not for: users expecting multi-model routing.

Use it when you want one stable endpoint. Skip it if you need video support.

Common questions

What is dolphin uncensored and how does it work?

It is an OpenAI-compatible chat-completions API serving one uncensored large language model. You send text in and get text out via POST /v1/chat/completions. The model id is 'uncensored' and runs on our servers without multi-model routing.

Does the API support streaming responses?

Yes. The API supports streaming via Server-Sent Events (SSE). You receive token usage information in the last chunk of the stream. This allows real-time token display in your client applications without waiting for the full completion.

What is the context window size?

The context window is 100,000 tokens for prompt and completion together. The max output is 16,000 tokens per request, or 2,048 tokens if you do not set max_tokens. This handles long documents efficiently without truncation errors.

Can I use this with the standard OpenAI SDK?

Yes. Change the base_url to HTTPS://api.dolphinuncensored.com/v1 and provide your API key. The endpoint structure matches the official OpenAI SDK standards. You can drop in our URL and immediately access the uncensored model with existing code.

How much does the API cost per token?

Input tokens cost $0.15 per 1 million tokens. Output tokens cost $0.50 per 1 million tokens. Prepaid credit is charged by real usage. Errors and refusals do not consume your credit balance, keeping costs predictable.

Which payment methods do you accept?

Crypto only: USDT on TRC20 or USDC on Base, any whole amount from $20 to $500. Bonus credit applies from $50.

Get your key in one step

Try with trial credit