DEVELOPER CENTER

Connect ideas to production.

API references, setup guides, and smart technical support. Get a key in 60 seconds and run your first request in three steps.

Setup guideEnvironment variablesCodex guideFAQ
60 SECOND START

Connect in three steps.

01

Get an API key

Create a key in the Tokenly dashboard and use it across four protocol surfaces.

View next step →
02

Set BASE_URL

Point requests to https://api.tokenly.ai/v1 and keep your familiar OpenAI SDK.

View next step →
ONE-CLICK CONFIGURATION

Configure your desktop AI app in four steps.

Create a token, install CC Switch, choose the app/token/model, allow the browser handoff, and restart your desktop app.

1

Create an API token

Open Token management, choose a custom name and provider group, then set optional expiry or quota limits.

2

Download and install CC Switch

Open CC Switch setup, download the Windows installer, install it, and keep the app running.

3

Select app, token, and model

The model list follows the selected token provider, so Gemini tokens show Gemini models and Claude tokens show Claude models.

4

Allow, save, and restart

Click Configure, choose Allow when prompted, save in CC Switch, then restart the desktop AI application.

READY TO TRY IT?

Open the live configuration workspace.

The setup page previews the provider, endpoint, masked key, and model before handoff.

Open CC Switch ↗
REFERENCE MAP

Everything has a direct path.

Use these links to move between the catalog, account, token controls, and logs without dead ends.

Token managementCreate, edit, copy, disable, and delete API tokens.Models & pricingCompare provider pricing, context windows, and capabilities.Usage logsReview every request, model, token, input, output, and cost.Task logsTrack configuration and asynchronous workspace events.DashboardOpen the full workspace overview and clickable metrics.
OPENAI-COMPATIBLE API

Familiar interfaces, minimal migration.

Compatible with mainstream OpenAI SDKs, plus Claude Messages and Gemini Generative.

Endpointhttps://api.tokenly.ai/v1
AuthenticationBearer TOKENLY_API_KEY
Models14 active models
Rate limitsPer-account and per-project quotas
ErrorsStandard HTTP + request_id
SDKscURL · Python · JavaScript
# Replace only the model name
curl https://api.tokenly.ai/v1/chat/completions \
  -H "Authorization: Bearer $TOKENLY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"gpt-5.6-sol","messages":[{"role":"user","content":"Hello"}]}'