Drop-in OpenAI Compatibility
Uses the standard /v1/chat/completions endpoint. Switch your base_url and keep your existing client code working immediately.
One key. Uncensored. OpenAI compatible.
An independent guide to the minimax api ecosystem and a drop-in, open-weight alternative for unrestricted text generation.
Try it in one request
curl https://api.minimaxapikey.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."}]
}'from openai import OpenAI
client = OpenAI(base_url="https://api.minimaxapikey.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.minimaxapikey.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);Uses the standard /v1/chat/completions endpoint. Switch your base_url and keep your existing client code working immediately.
The model answers without refusing lawful adult, controversial, or security-research topics. Only minor sexual content is blocked.
Process up to 100,000 tokens per request. Ideal for long documents or extended conversation histories without losing context.
Pay only for what you use. Prepaid credit never expires. No monthly subscriptions or hidden egress fees.
Your prompts are not used to train the model. No phone number or credit card is required to start testing.
One key per account. Regenerate instantly if needed. Supports streaming and function calling for complex integrations.
Sign up with just an email address to receive your first API key instantly.
Point your OpenAI-compatible SDK to https://api.minimaxapikey.com/v1 and inject your key.
Send a chat completion request to the uncensored model and receive unrestricted text output.
Generate stories with adult themes or complex characters without the model refusing based on content filters. Perfect for novelists and roleplayers who need raw output.
Test LLM vulnerabilities and prompt injections without the model refusing to answer controversial questions. Get direct, unfiltered responses for analysis.
Feed entire books or technical manuals into the 100k context window. Extract insights or summarize long texts without chunking overhead.
Rapidly prototype chat interfaces using standard OpenAI SDKs. The uncensored mini max api behaves like other major models but with fewer restrictions.
Many production applications fail not because of accuracy, but because of refusals. Standard models often block lawful adult content, controversial opinions, or specific creative directions. The minimax api we host serves an open-weight model tuned to answer these queries directly. This is critical for apps where the user expects the model to follow instructions without moralizing or shutting down.
We do not offer image, audio, or video capabilities. We focus solely on high-quality text generation. If your use case requires multimodal outputs, you will need a different provider. For text-only workflows, our uncensored model provides consistent, filter-free results.
Our API implements the standard /v1/chat/completions endpoint. You can use the official OpenAI SDKs for Python, Node.js, or any other language. Simply update the base_url to https://api.minimaxapikey.com/v1 and provide your API key. The model ID to use is uncensored.
This compatibility means you can swap in our service without rewriting your application logic. It works with tools like LangChain, LlamaIndex, and custom clients. We support streaming via Server-Sent Events and function calling for dynamic tool use.
Note: We are not an official partner of OpenAI. We provide a compatible interface, not the GPT models themselves.
Most standard models offer 8k or 32k tokens. Our uncensored model supports a 100,000 token context window. This allows you to pass entire documents, codebases, or long conversation histories in a single request. It reduces the need for complex chunking and retrieval strategies.
The context includes both the input prompt and the output completion. This makes it ideal for summarizing long texts or maintaining state in complex multi-turn conversations. You can send up to 8 MB per request body.
We charge based on actual usage. Input tokens cost $0.15 per 1 million tokens. Output tokens cost $0.50 per 1 million tokens. There are no monthly subscriptions or commitment fees. You prepay credit, and it never expires.
You can top up from $20 by crypto (USDT on TRC20, USDC on Base). Bonus credit is added for larger deposits: +5% at $50, +10% at $100, and +15% at $200. This model ensures you only pay for what you use, making it cost-effective for both small tests and high-throughput production apps.
Your data is not used to train our model. When you send a request, the text is processed to generate a response, but it is not stored for training purposes. This is a key difference from some major cloud providers that use customer data to improve their base models.
We require only an email and password to create an account. No phone number or credit card is needed for the trial. You can regenerate your API key at any time, revoking the old one instantly. This gives you full control over your access credentials.
New accounts receive $0.50 of trial credit valid for 7 days. Claim this credit with one click on your API key page. No credit card is required to start. This allows you to test the API, verify compatibility, and evaluate the uncensored output quality before committing to a purchase.
Each person can claim one trial. Once the trial credit is used or expires, you can top up with your own funds. This low-friction entry makes it easy to evaluate the uncensored mini max api for your specific use case.
No, this is an independent service. We host an open-weight model and provide an OpenAI-compatible interface. We are not affiliated with the original creators of the MiniMax model or any other vendor.
Use the model ID "uncensored" in your API requests. This identifies the specific open-weight model hosted on our servers.
Yes, streaming is supported via Server-Sent Events (SSE). You can set the stream parameter to true in your request to receive tokens as they are generated.
No. We do not use your prompt data to train our models. Your data is processed to generate responses but is not stored for training purposes.
You are limited to 300 requests per minute per API key. The maximum request body size is 8 MB. These limits ensure stable performance for all users.
Create an account, copy the key, change the base URL. That is the whole setup.
Practical articles on setup, pricing and integration.
MiniMax API in Python: Use the OpenAI SDKMiniMax API Python guide: use the official openai package with a custom base URL for chat, streaming, JSON mode, tool calls, retries and cost…
OpenAI Alternative to Claude Code: Coding Agent SetupLooking for an OpenAI alternative to Claude Code? Point OpenAI-compatible coding tools like Aider or Continue at an uncensored text API with one…
MiniMax API Setup: Key, Base URL and First RequestMiniMax API setup in six steps: sign up, claim the trial, create a key, set the base URL, send a curl request and fix common errors. No card…
Groq API: the options comparedUse MiniMax as an OpenAI-compatible Groq alternative. Pay per token for uncensored access. Get your API key now.