Quickstart

Connect in minutes.

Create a key in your dashboard, add funds using an available payment method, and use the compatibility mode your application already speaks — or drop the base URL into any of the tools below.

If you have only used ChatGPT before

UndercutAI is not a separate chat website. It is API access to AI models that you connect to familiar apps such as Claude Code, Cursor, Cline, n8n, or your own code. Choose an app, enter our URL and API key, and it sends requests through UndercutAI.

How this differs from ChatGPT

ChatGPT is a ready-made interface. Here, you pay for and connect API access to a third-party tool or your application. ChatGPT Plus is not required.

⚡Interactive Setup

Inject your credentials into snippets

Get key in dashboard →

Enter your API key and choose a model — they will be automatically injected into every code block and command below. Stays inside your browser tab.

Your API Key

Default Model

Quick Base URL:

Select tool or framework

⚡

Cursor

Native

Connect via native OpenAI Compatible mode. Works seamlessly in chat and file edits.

1Open Settings → Models (or press Ctrl/Cmd + Shift + J).

2Under OpenAI API Key, set Base URL to https://undercut.pro/v1 and enter your key.

3Click Add model and enter your chosen model identifier.

💡You can add multiple model identifiers from our table and switch between them inside Cursor chat.

Ready-to-use snippet / command

// Cursor Settings → Models → OpenAI API Key
Base URL:  https://undercut.pro/v1
API Key:   gw_your_key
Model:     claude-opus-5

Three terms you will see below

API key: the password that lets an application use your balance.
Base URL: the address of the UndercutAI service.
Model: the specific AI model that will answer your request.

Keep your API key private

Never publish it, commit it to Git, or send it in a chat. Anyone with the key can spend your balance.

Model names

These models support context windows up to 1 million tokens through UndercutAI, except Grok, which supports up to 600,000 tokens. Use an identifier as the model parameter, as ANTHROPIC_MODEL, or in your client's model switcher (for example /model claude-sonnet-5 in Claude Code). Per-token prices are on the pricing table.

Model	Identifier
Claude Opus 5	`claude-opus-5`
Claude Sonnet 5	`claude-sonnet-5`
Claude Fable 5	`claude-fable-5`
Claude Fable 5.1	`claude-fable-5.1`
GPT-6 Astra	`gpt-6-astra`
GPT-5.6 Sol	`gpt-5.6-sol`
GPT-5.6 Terra	`gpt-5.6-terra`
GPT-5.6 Luna	`gpt-5.6-luna`
Gemini 3.1 Pro	`gemini-3.1-pro-preview`
Gemini 3.7 Flash	`gemini-3.7-flash`
Gemini 3.8 Flash	`gemini-3.8-flash`
Grok 4.6	`grok-4.6`
Kimi K3	`kimi-k3`

Batch API availability

A public Batch API is not currently available. Ordinary requests use the published token rates. Adding a custom batch header or sending several messages does not enable batch processing or a discount.

Cost calculator

Estimate the cost of a request before you send it.

ModelInput tokensOutput tokens

$5.0625

Standard token estimate without cache reads or writes. Actual cost depends on the usage reported by the gateway.

Billing note

Dashboard costs use the configured rates for ordinary input, cache reads, cache writes and output. Cache categories reported by the gateway are removed from ordinary input to avoid double billing. The applied rates and pricing version are saved with each usage record; updating the price list does not rewrite recorded charges.