English简体中文
freebrain/v1

Up to −92.5%vs official list prices

One API for all your models.workflows.ideas.

Connect to a growing catalog through one compatible endpoint.

Get an API key →
export OPENAI_BASE_URL=https://api.thefreebrain.com/v1
point at a module

Models in the catalog from

01/Pricing

Same models. Less spend.

Use Discount Offers to lower your AI costs: up to 92.5% off list price on selected text models. Each model’s own rate is in the table below.

Price snapshot: 2026-10-08. Input/output prices are per million tokens. Discounts vary by model and billing item. Confirm current rates in the model plaza.

02/Layer

Everything between your code and the model.

Use one compatible API for every model currently live, with per-key controls, visible usage and pay-as-you-go billing. Add new models without rebuilding your integration.

21

models in the catalog

2

model providers

4

configured API endpoints

1.05M

tokens max published context

Fig. 01One strip, a socket for every format.

One API, multiple compatible endpoints

Connect with leading AI clients. Check the model plaza for each model’s supported endpoints and the API docs for integration details.

Fig. 02A key for each job, each with its own limit.

Clear per-key controls

Set quotas, spend caps and rate limits per key, then review usage and billing in the console.

Fig. 03Every request prints a line.

Every request stays measurable.

From request to model, usage and bill, every step is recorded in the console so the actual cost never has to be guessed.

Fig. 04The switch shows where each request goes.

Routing you can see.

Each request goes through one gateway to a configured channel, while the model ID, token usage and billed amount stay visible.

03/Developers

Connect once. Switch models freely.

Keep your existing client. Change the base URL and model ID, then move between models without rebuilding your stack.

Works with Cherry Studio, CC Switch and apps that support leading AI API protocols.

Read the API docs

Claude Code setupCodex setupMigration checklistPython SDKNode.js SDK

Fig. 05Same terminal. New base URL.
python 
from openai import OpenAI client = OpenAI(base_url="https://api.thefreebrain.com/v1", api_key="YOUR_API_KEY")reply = client.chat.completions.create(model="gpt-6.1-sol",    messages=[{"role": "user", "content": "Say hello from the gateway."}])print(reply.choices[0].message.content)

Example model: gpt-6.1-sol. See the model plaza for rates and endpoint availability.

04/Quick start

Pick your tool. Make your first call in a minute.

Paste an API key and every command, config file and one-click import below picks it up. The key stays on this page: it is never uploaded or stored.

No key yet: commands show YOUR_API_KEY and one-click imports stay off.

No key yet? Create one in the console →

Fig. 06One key to paste. Nothing else to type.
01

One-line command

Paste into a terminal. It applies to this run only, good for a first try.

ANTHROPIC_BASE_URL=https://api.thefreebrain.com ANTHROPIC_AUTH_TOKEN=YOUR_API_KEY claude
02

Save to settings

~/.claude/settings.json

Merge into the env block of this file, then just run claude. Opus, Sonnet and Haiku map to models that are online.

{
  "env": {
    "ANTHROPIC_BASE_URL": "https://api.thefreebrain.com",
    "ANTHROPIC_AUTH_TOKEN": "YOUR_API_KEY",
    "ANTHROPIC_DEFAULT_OPUS_MODEL": "claude-opus-5-5",
    "ANTHROPIC_DEFAULT_SONNET_MODEL": "claude-sonnet-5-5",
    "ANTHROPIC_DEFAULT_HAIKU_MODEL": "claude-haiku-4-5-20251001"
  }
}
03

Import to CC Switch

With CC Switch installed, add FreeBrain and make it Claude Code's provider in one click.

Add your key above first.

One-click imports need the app installed on this computer. On Windows PowerShell, set variables with $env:NAME="value" instead of export.

05/For everyone

From a side project to production.

The same low-cost gateway can carry you from the first request to a stable production workload.

Fig. 08A weekend project…

Solo developers

Ship faster, pay per use.

  • One API key for every live model, with no separate vendor accounts.
  • Drop-in compatible with leading AI API clients.
  • Metered billing with usage and spend visible live.
  • No gateway infrastructure to run; start in minutes.

Create an API key

Fig. 09…or a key per team, each behind its own door.

Enterprise teams

Control and reliability at scale.

  • Set quotas and spend caps per key.
  • Apply rate limits for different teams and projects.
  • Review usage, cost and latency per request.
  • Add channels later without changing client code.
  • Confirm every live model price before production use.

Talk to our team →

Three steps

  1. 01

    Configure

    Create a key in the console, top up balance and set a quota.

  2. 02

    Connect

    Point any compatible client at the gateway base URL.

  3. 03

    Monitor

    Track usage, cost and latency live, per key and request.

06/Start

Start with one key.

Create a key, paste it into the client you already use, and send your first request.

Start building

Fig. 10One connection left to make.