Tokens: API keys and access control in OpenOpen8

A token is an API key that you create in the OpenOpen8 dashboard. You include it in every request you send to OpenOpen8, and OpenOpen8 uses it to identify you, check your permissions, and track your usage.

Authentication

Pass your token in the Authorization header as a Bearer token:

Authorization: Bearer YOUR_TOKEN

This matches the standard OpenAI client convention, so any existing client library or tool that accepts an OpenAI API key works with OpenOpen8 tokens without modification.

cURL
Python (OpenAI SDK)
Node.js (OpenAI SDK)

curl https://openopen8.ai/v1/chat/completions \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"model": "gpt-4o", "messages": [{"role": "user", "content": "Hello"}]}'

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_TOKEN",
    base_url="https://openopen8.ai/v1"
)

response = client.chat.completions.create(
    model="gpt-4o",
    messages=[{"role": "user", "content": "Hello"}]
)

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "YOUR_TOKEN",
  baseURL: "https://openopen8.ai/v1",
});

const response = await client.chat.completions.create({
  model: "gpt-4o",
  messages: [{ role: "user", content: "Hello" }],
});

Token properties

Property	Description
Name	A label for your own reference. Useful for tracking which application or team uses each token.
Quota	The maximum number of credits this token can consume. Set to unlimited for no cap.
Expiry date	The date after which the token is no longer accepted. Leave blank for no expiry.
Allowed models	A list of model names this token is permitted to request. If empty, the token can access all models available to its group.
Allowed IPs	A comma-separated list of IP addresses or CIDR ranges allowed to use this token. Leave blank to allow any IP.
Group	The group this token belongs to. Groups determine pricing ratios and access scopes.

Quota and usage

OpenOpen8 tracks quota consumption for each token. Usage is calculated in credits based on the model’s pricing ratio — models with higher pricing consume more credits per token of input or output.

Model pricing ratios are configured by OpenOpen8. You can view the effective rate for each model in the dashboard.

When a token’s quota is exhausted, OpenOpen8 returns an error and rejects further requests until the quota is refilled or raised. You can monitor current usage from the Token details page in the dashboard.

Model restrictions

You can lock a token to a specific set of models. For example, you might create a token that can only call gpt-4o-mini to control costs, or one that can only call claude-3-5-sonnet-20241022 for a specific workflow.

Model restrictions are enforced before routing. If a request specifies a model not on the token’s allowed list, OpenOpen8 rejects it immediately without consuming any quota.

Groups

Every token belongs to a group. Groups control pricing ratios and access scopes:

Different groups can have different pricing multipliers
Groups segment traffic and access permissions

This lets you manage different tiers — for example, giving one team standard pricing while offering another team a discounted rate.

Manage tokens

Step-by-step guide to creating and managing tokens.

Authentication reference

Full details on the Authorization header and supported formats.

Get Started

Basics

Guides

Help

Tokens: API keys and access control in OpenOpen8

Authentication

Token properties

Quota and usage

Model restrictions

Groups

Manage tokens

Authentication reference

Get Started

Basics

Guides

Help

​Authentication

​Token properties

​Quota and usage

​Model restrictions

​Groups

​Related

Manage tokens

Authentication reference

Authentication

Token properties

Quota and usage

Model restrictions

Groups

Related