Welcome to TokenReply
TokenReply is an API made for roleplay. Every model has a Roleplay score and a Value score, and some models are free. To connect an app, you need the API URL, your API key and a model name.
What you get
- Models from OpenAI, Claude, Gemini, Llama and more, all with one API key
- Cache prices: your app resends the story with every message, and repeated history is billed at the cache price on models that have one
- Pay from your usage allowance or wallet, and cap spending on each API key in USD
- The OpenAI format: Chat Completions, Responses, Embeddings, Images, Audio and more
- Streaming over SSE, so replies appear as they're written
- A higher rate limit on Plus
Quick start
Get started in three steps:
- Create an account with your email or another sign-in option
- Create an API key on the API Key page and give it a name
- Connect your app with the API URL https://api.tokenreply.com/v1, your key and a model name
Your first month
New accounts get a one-month Trial with a starter usage allowance. To keep going after that, check in daily, top up your wallet or upgrade to Plus.
All guides
Roleplay apps
- SillyTavern setup for the Chat Completion connection and recommended settings
- JanitorAI setup for the proxy settings and fixes for common errors
- Roleplay API to compare models by Roleplay and Value score
User guide
- Registration and sign-in for email sign-up, verification and other sign-in options
- Using API keys for the API URL https://api.tokenreply.com/v1 and examples in Python, Node.js and cURL
- API key management to create keys, set limits and replace old ones
- Plans and tiers to compare Default, Trial and Plus and see how to upgrade
- Billing and balance for what gets charged first, top-ups and your spending history
- Daily check-in for free usage allowance every day
- Referral rewards to earn allowance and free Plus time by inviting friends
API reference
- Base URL and authentication for URLs, headers and request formats
- Compatibility for supported endpoints and the apps you can connect
- Models and rules for which models you can use, their IDs and prices
AI coding tools
- Install Node.js, the runtime these coding tools need
- Claude Code setup, Anthropic's coding tool for the terminal
- Codex setup, OpenAI's coding assistant
- Gemini CLI setup, Google's coding assistant
Tools and apps
- CC-Switch to switch coding tool settings from one desktop app
- Third-party apps for Cherry Studio, OpenCode and other apps
Help
- Code examples for streaming, direct HTTP and embeddings
- FAQ and troubleshooting for common errors and fixes
- Models for every model and its price
