UNLI LogoUNLI

Never pay per token again,
flat pricing with unlimited tokens.

Unlock unlimited usage across GPT, Gemini, DeepSeek, Qwen, GLM, Kimi and more models.

Get Started
JavaScriptexample.js
import OpenAI from "openai";

const openai = new OpenAI({
  baseURL: "https://api.unli.dev/v1",
  apiKey: "YOUR_UNLI_API_KEY",
});

const completion = await openai.chat.completions.create({
  model: "auto",
  messages: [{ role: "user", content: "Hello World!" }],
  stream: false
});

console.log(completion.choices[0].message.content);

186 tokens or 48,210.Both count as 1 request.

Unlimited tokens on every plan, free included.

  • Works with OpenAI and Anthropic SDKs
  • GPT, Gemini, DeepSeek, Qwen, GLM, Kimi and more
  • Streaming, tool calls and images
  • Prompts aren't stored
Start free

No credit card required.

Example on the Starter plan

4,997 of 5,000 requests left this month

Example requests and how each one is counted
RequestTokensCounts as
Quick question1861 request
Code review12,9041 request
Agent run on a large repo48,2101 request
Failed request0Not counted

Plans and Pricing

Hobby

Free

For your hobby projects

  • Unlimited tokens
  • 100 requests/month
  • 20 requests/minute

Starter

$5

Great for small applications

  • Unlimited tokens
  • 5,000 requests/month
  • 30 requests/minute

Pro
Popular

$15

Great for growing businesses

  • Unlimited tokens
  • 20,000 requests/month
  • 100 requests/minute

Ultra

$125

For large-scale applications

  • Unlimited tokens
  • 200,000 requests/month
  • 200 requests/minute

Enterprise

Custom

Custom for your needs

  • Unlimited tokens
  • Custom requests/month
  • Custom requests/minute

Frequently Asked Questions

Yes. Every plan has unlimited tokens, including the free Hobby plan. Plans differ in how many requests you can send.

Each API call counts as one request, however many tokens it uses. Failed requests are not counted.

No credit card required. You only add a payment method when you upgrade to a paid plan.

Yes. UNLI supports the OpenAI Chat Completions format at https://api.unli.dev/v1 and the Anthropic Messages format at https://api.unli.dev, so you only change the base URL and API key.

We support a wide range of models including GPT, Gemini, DeepSeek, Qwen, GLM, Kimi, and more.

No. We do not log any of the prompts or completions sent to our API.

No, remaining requests do not roll over to the next month.

Stop counting tokens.

Access unlimited GPT, Gemini, DeepSeek, Qwen, GLM, Kimi and more.
Every request counts as one, however many tokens it uses.

Start freeContact us