DeepSeek jailbreak API documentation

Authentication and base URL

Start by setting your base URL to https://api.deepseekjailbreak.com/v1 and passing your API key in the Authorization header. This endpoint works with the official OpenAI SDKs and any compatible client. You get one active key per account; generating a new key replaces the old one immediately. Sign up via Google or email to retrieve your key instantly without entering a phone number.

Copycurl https://api.deepseekjailbreak.com/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "uncensored",
    "messages": [{"role": "user", "content": "Explain how the context window works in this model"}]
  }'

Keep your key secure since it grants full access to your prepaid credit. The API rejects requests with invalid or expired headers using standard HTTP codes.

Sending your first chat completion

Send a POST request to /v1/chat/completions with your message and the model ID uncensored. This model is tuned to answer without refusing lawful adult, fictional, or controversial topics. It is an open-weight model running on our servers, not a proxy for GPT or DeepSeek. You can send up to 8 requests simultaneously per key.

Copyfrom openai import OpenAI

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

Each request consumes input tokens from your prepaid balance. Errors and refusals do not charge you. The model supports a context window of 100,000 tokens for both prompt and completion combined.

Streaming responses and token usage

Enable streaming by setting stream: true to receive tokens via Server-Sent Events (SSE). This is useful for real-time UI updates or long-form content generation. Token usage data appears in the final chunk of the stream, allowing you to track costs accurately. Streaming works with all standard parameters like temperature and top_p.

Copyimport OpenAI from "openai";

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

Use this approach when you need immediate feedback rather than waiting for the full response. The API handles partial token delivery efficiently.

Configuring temperature and top_p

Adjust temperature to control randomness and top_p for nucleus sampling. Lower values make outputs more deterministic, while higher values increase creativity. These parameters help fine-tune the uncensored model for specific tasks like coding or creative writing. You can also set seed for reproducible results.

The model supports stop sequences and penalty parameters like presence_penalty. Use these to control output length and repetition. Precise configuration ensures consistent behavior across runs.

Using function calling and JSON mode

Enable function calling by defining tools in your request. The model supports structured outputs for reliable integration. Use response_format: {"type": "json_object"} to force JSON output. This is ideal for parsing data or generating structured responses. The API ensures strict adherence to the specified schema.

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)

Combine streaming with JSON mode for real-time structured data. This setup works well for dynamic applications requiring predictable formats. Always validate the output structure in your client code.

Handling errors and rate limits

Handle HTTP 401 for invalid keys, 402 for insufficient credit, and 429 for rate limits. The API allows 300 requests per minute per key. You can send up to 8 concurrent requests. Request bodies must not exceed 8 MB. Errors do not consume your prepaid credit.

Implement exponential backoff for 429 responses. Monitor your credit balance to avoid service interruption. The API provides clear error messages to help debug issues quickly.