EN ▾

Uncensored OpenAI-compatible API: a Claude Code alternative

Claude Code Proxy alternative you can call with the same client code

An uncensored OpenAI-compatible endpoint with one model of its own. Swap the base URL and API key to access a single, robust model without complex local setups.

  • Drop-in OpenAI compatibility
  • 100k token context window
  • No prompt training privacy

Get API keyRead the docs

Try it in one request

curl https://api.claudecodeproxyapi.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."}]
  }'

Not Claude and not affiliated with Anthropic: one uncensored open-weight model, OpenAI-compatible API, one price. Claude Code can't connect directly: it needs a local translator such as LiteLLM.

per 1M input tokens
$0.15
Output tokens / 1M
$0.50
token context
100,000
trial credit
$0.50
requests per minute
300

What you get

Single Model Focus

Unlike multi-vendor routers, we optimize for one uncensored open-weight model. This ensures consistent latency and predictable behavior for your coding workflows.

True Uncensored Output

The model answers controversial, adult, and security-research topics without standard refusals. It blocks only sexual content involving minors, allowing lawful adult use.

Standard OpenAI Endpoint

Use the official SDKs with our base URL. Streaming, tool calling, and chat completions work exactly as expected with the standard OpenAI-compatible format.

Transparent Pricing

Pay-as-you-go with prepaid credit that never expires. Rates are $0.15 per 1M input tokens and $0.50 per 1M output tokens, with bonuses for larger top-ups.

Privacy First

Your prompts are not used for training. Sign up with just an email and password; no phone number or credit card is required to start.

High Rate Limits

Each API key supports 300 requests per minute and 8 MB request bodies. Keys can be regenerated instantly if compromised, revoking the old one immediately.

How it works

Generate Your Key

Sign up with email and password to receive your unique API key instantly.

Update Configuration

Point your client to https://api.claudecodeproxyapi.com/v1 and set your new key.

Start Prompting

Send requests using the standard /v1/chat/completions endpoint with model ID 'uncensored'.

What people build with it

Creative Writing

Generate stories and scripts without content filters blocking mature themes or stylistic choices. The model handles nuanced adult narratives without unnecessary refusals.

Security Research

Analyze code and vulnerabilities without the model hedging on technical risks. Get direct, uncensored analysis of security patterns and exploit details.

Data Annotation

Process large datasets with consistent formatting. The single-model focus ensures uniform output quality for labeling tasks without routing variability.

Dev Assistant

Connect Claude Code through a local translator to a model that doesn't over-explain or refuse standard technical queries. Maintain your existing workflow with improved response freedom.

Why a dedicated API as a Claude Code alternative?

Most LLM gateways route traffic across dozens of vendors, introducing latency and inconsistent behavior. This proxy is built for developers looking for a Claude Code alternative who need a reliable, high-performance endpoint. By focusing on a single uncensored open-weight model, we eliminate the variability of multi-vendor routing. You get direct access to a model tuned for responsiveness and depth, running on our own GPU servers. This specialization means lower latency and more predictable token usage compared to general-purpose routers that share infrastructure across many clients.

Drop-in OpenAI Compatibility

Your existing code works without modification. We expose a standard OpenAI-compatible endpoint at https://api.claudecodeproxyapi.com/v1. Simply update your client's base URL and API key. The endpoint supports streaming via Server-Sent Events (SSE) and function/tool calling. The model ID is 'uncensored'. This compatibility extends to the official OpenAI SDKs and any client that speaks the OpenAI protocol. You retain full control over your prompts and parameters while benefiting from our optimized infrastructure.

curl https://api.claudecodeproxyapi.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."}]
  }'

Uncensored & Lawful Adult Use

The model does not refuse lawful adult, fictional, or controversial topics. It answers directly without the typical hedging or moralizing found in gated models. The only hard limit is on sexual content involving minors, which is always blocked. This makes it ideal for creative writing, roleplay, and technical analysis where content filters might otherwise interrupt the flow. It is not GPT, Claude, Gemini, or any other vendor model; it is an open-weight model trained for unfiltered responses.

Who This Is Not For

This API is strictly for text input and text output. We do not offer embeddings, image generation, audio, video, or fine-tuning. If you need multi-modal capabilities or model routing between different providers, this single-model proxy is not the right fit. We also do not offer SLA guarantees, SOC2 certifications, or on-prem deployment options. It is designed for developers who want a simple, uncensored text API without enterprise compliance overhead or complex configuration.

Questions and answers

Is this the official Claude API?

No. This is an independent proxy for an uncensored open-weight model. It is not GPT, Claude, Gemini, or any other vendor model. It is compatible with the OpenAI API format but runs on our own infrastructure.

How much does it cost?

Pricing is $0.15 per 1M input tokens and $0.50 per 1M output tokens. You pay for what you use with prepaid credit that never expires. There are no monthly fees or subscriptions.

Do you use my prompts for training?

No. We do not use your prompt data for training. You only need an email and password to sign up, and your data remains private.

Can I stream responses?

Yes. The /v1/chat/completions endpoint supports streaming via Server-Sent Events (SSE). You can also use function calling and standard chat completions.

What are the rate limits?

You get 300 requests per minute per key and an 8 MB request body limit. You can generate a new key at any time, which immediately revokes the old one.

Your key is one form away

Create an account, copy the key, change the base URL. That is the whole setup.