Venice AI API Documentation and Quickstart

Authentication and Base URL

Your venice ai api key authenticates every request to the API. Replace the standard base URL in your SDK configuration with the one above. The system uses prepaid credit, so ensure your balance is sufficient before sending heavy payloads.

Chat Completions Endpoint

Send text to the model using the standard chat completions endpoint. The model ID is always uncensored, ensuring consistent behavior without content refusals for lawful adult topics.

Configure your client to point to the correct base URL and include the model identifier in the request body. You can adjust parameters like temperature or top_p to control creativity.

Errors and refusals are free, so you only pay for actual token usage. The context window supports up to 100,000 tokens for both input and output combined.

Streaming and SSE Usage

Enable real-time token generation by setting stream: true in your request. The API returns Server-Sent Events (SSE) that you can parse incrementally.

Each chunk contains partial text, but token usage metrics appear only in the final chunk. This approach reduces perceived latency for users waiting for long-form content.

Handle the stream end event to finalize the response. Ensure your client buffers the accumulated text correctly before presenting it to the end user.

Function Calling and JSON Mode

Use the API for structured data extraction or tool execution. Set response_format to {"type": "json_object"} to enforce strict JSON output.

Define tools in the request body to allow the model to call functions. The model will return structured arguments that your backend can execute.

This feature is ideal for building agents or automating workflows. Ensure your function definitions are clear to minimize hallucination in the arguments.

Parameters and Settings

Control the model's behavior using standard parameters. Use temperature for creativity, top_p for nucleus sampling, and stop sequences to truncate output.

Set seed for reproducible results during testing. Adjust presence_penalty and frequency_penalty to influence topic diversity.

The maximum output is 16,000 tokens per request. If you do not specify max_tokens, the default is 2,048. Plan your prompts to stay within the 100,000 token context window.

Rate Limits and Constraints

Each API key allows 300 requests per minute and 8 concurrent requests. The maximum request body size is 8 MB.

If you exceed these limits, you will receive a 429 error. A single account supports one active key; generating a new key replaces the previous one.

Credit never expires, so you can top up whenever needed. Use crypto for top-ups, with bonus credit available for larger amounts.

cURL

Copycurl https://api.veniceaiapikey.com/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "uncensored",
    "messages": [{"role": "user", "content": "Generate a summary of the current weather conditions"}]
  }'

Python

Copyfrom openai import OpenAI

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

Node.js

Copyimport OpenAI from "openai";

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

Streaming

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)