An uncensored AI for code, on a system you are allowed to touch

MadGPT is for the function, the patch, the test, the review, and the security work on a system you own or are allowed to test. You paste the code. It answers.

Coder and v4.1 are where a patch, a port, a failing test, or a language you do not write every day should land. v5.1 is the flagship when you picked the hard problem on purpose. MadGPT Uncensored is the paid reasoning slot for the lawful question a filtered assistant leaves, and only a paid plan can open it. Auto will not put that question on MadGPT Uncensored. You pick the slot yourself. The answer is code, with the mistakes a model makes. A patch you are about to ship needs a check outside the chat. The model does not run your tests, and it does not know a system it has not seen.

Security research sits next to that work. The work is penetration testing guidance, vulnerability analysis, and exploit development on a system you own or are allowed to test. A request to break into someone else's network is outside the product. The same sentence covers a scanner pointed at a host you do not administer, a payload for an account that is not yours, and a write-up whose only use is the break-in. If the system is yours, or the owner told you to test it, that is the job the coding slots are for.

A filtered assistant often spends the reply explaining why it will not look at the crash, the proof of concept, or the ugly part of a protocol. MadGPT is pointed at those conversations when they are lawful. It can still miss the bug. Medicine and law are not this page. A production deploy is.

You can attach the file the job is about, or a screenshot of the stack trace. On 1 Month, 3 Months, and 1 Year that is 4 text or code files a day and 4 images a day. Lifetime: unlimited file and image uploads. Images can be jpg, jpeg, png, webp, or gif, up to 2 MB. MadGPT can read the screenshot. It does not turn a prompt into a picture.

The VS Code extension is included on Lifetime, and Agent Mode is part of that extension. Agent Mode is not a mode of the web chat.

The plans are 1 Month at $99, 3 Months at $250, 1 Year at $999, and Lifetime at $2499.

Generations on those plans are unlimited.

Web search is included.

Credit packs are a separate purchase, and they do not include uploads.

Crypto at checkout is BTC, ETH, SOL, and USDT, plus a manual payment in Telegram.

Registration does not add credits.

The signup mail does not contain a key.

The Telegram bot issues a trial key for 7 credits.

The 7 days start at the first chat.

Paste the key on the site.

No credit card.

The trial cannot open MadGPT Uncensored.

Chats are stored on your account so the history is there when you return.

Deleting the account deletes that copy.

Chats are sent to third-party model providers to produce the reply, and those providers have their own retention policies.

This app does not run its own training job on them.