Skip to main content
4096bytes gives developers an AI model access layer that fits real coding workflows. Create an API Key in the dashboard, then import it into clients such as Codex, OpenCode, OpenClaw, Claude Code, Cursor, and Cherry Studio, or connect it to your own workflow through Codex SDK.
Use the Base URL, model name, and API Key shown in the dashboard. This documentation explains where to configure each value, so you do not copy stale settings.

​
Who it is for

​
Start here

Create an account

Sign in to the dashboard and prepare to manage plans, balance, and API Keys.

Get an API Key

Create a client API Key and learn how to store it safely.

Supported models and pricing

Review current models, input and output costs, cache pricing, and discount notes.

​
Connect common clients

If you already have an API Key, open the client page you need.

CC Switch

Import 4096bytes into CC Switch and manage CLI providers in one place.

Codex

Configure 4096bytes in Codex for terminal and local development tasks.

Codex Desktop

Reuse the Codex CLI config in the Codex desktop app.

Codex VS Code extension

Reuse the local Codex CLI config in VS Code.

OpenCode

Write the dashboard-generated Provider config into OpenCode.

OpenClaw

Configure 4096bytes as a custom model Provider in OpenClaw.

Claude Code

Use the Claude Code compatible settings shown in the dashboard.

Cursor

Configure a custom provider and model in Cursor.

Cherry Studio

Add 4096bytes as a provider and test the connection.

​
Connect SDKs

Codex SDK

Call Codex from your application, internal tool, or automation workflow.

​
Troubleshooting

If configuration fails, start with FAQ. If the issue is still unresolved, see Contact us for the information to include.