Claude CodeKilo CodeZCodeompLangChain
CodexOpenHandsCommand CodeHermes Agentclaude-mem
OpenClawZed EditorRoo CodeDeepSeek HarnessSillyTavern
Open WebUIQwen CodegooseZoo CodeHammerAI
ClineCursorpiStrixJanitor AI

Connect the AI tools you already use to more models.

Find setup guides for Claude Code, Cline, OpenClaw, and more. Create a TokenHub API key, add your API base and model, and start using it.

View all apps

Start with common apps that have a clear setup path and make your first request in minutes.

Claude CodeAnthropic’s command-line coding agent for understanding codebases, editing files, running commands, and completing complex development tasks.CodexOpenAI’s coding agent for working with code, editing files, and running tasks from the terminal or development environment.OpenClawAn open-source personal agent that connects messaging apps, browsers, and local tools, with model switching through compatible APIs.Open WebUIA self-hosted chat and knowledge interface that can connect to any OpenAI-compatible model service.ClineAn open-source coding agent that runs in your editor and supports custom OpenAI-compatible endpoints and models.Kilo CodeAn open-source coding agent for VS Code, JetBrains, and the CLI with custom models and compatible API providers.OpenHandsAn open-source software agent that can run commands, browse the web, and edit code.Qwen CodeAn open-source coding agent for terminal workflows with configurable compatible model providers.CursorAn AI-native code editor. Custom models and Base URL support depend on your client version and workflow.

Browse all apps

Showing 25 apps

Claude CodeAnthropic’s command-line coding agent for understanding codebases, editing files, running commands, and completing complex development tasks.Anthropic CompatibleCodexOpenAI’s coding agent for working with code, editing files, and running tasks from the terminal or development environment.OpenAI CompatibleOpenClawAn open-source personal agent that connects messaging apps, browsers, and local tools, with model switching through compatible APIs.OpenAI / AnthropicOpen WebUIA self-hosted chat and knowledge interface that can connect to any OpenAI-compatible model service.OpenAI / AnthropicClineAn open-source coding agent that runs in your editor and supports custom OpenAI-compatible endpoints and models.OpenAI CompatibleKilo CodeAn open-source coding agent for VS Code, JetBrains, and the CLI with custom models and compatible API providers.OpenAI CompatibleOpenHandsAn open-source software agent that can run commands, browse the web, and edit code.OpenAI CompatibleZed EditorA high-performance collaborative code editor with custom OpenAI-compatible model endpoints.OpenAI / AnthropicQwen CodeAn open-source coding agent for terminal workflows with configurable compatible model providers.OpenAI CompatibleCursorAn AI-native code editor. Custom models and Base URL support depend on your client version and workflow.OpenAI CompatibleZCodeA coding agent that can add providers and configure OpenAI- or Anthropic-compatible models.OpenAI / AnthropicCommand CodeA BYOK command-line coding tool that can add OpenAI- or Anthropic-compatible providers.OpenAI / AnthropicRoo CodeAn open-source autonomous coding agent for VS Code with OpenAI-compatible providers.OpenAI CompatiblegooseAn open-source, extensible local agent with custom model providers.OpenAI / AnthropicpiA command-line coding agent built for extensibility and personal workflows.OpenAI / AnthropicompA terminal agent that supports custom OpenAI- and Anthropic-compatible providers.OpenAI / AnthropicHermes AgentAn open-source personal agent with persistent memory, tool use, and skills.OpenAI CompatibleDeepSeek HarnessAn open-source agent harness for DeepSeek models that requires self-hosting and API configuration.OpenAI CompatibleZoo CodeAn open-source coding agent with configurable OpenAI-compatible provider URLs, keys, and models.OpenAI CompatibleStrixAn open-source application security agent that supports LiteLLM or a custom API base.OpenAI CompatibleLangChainA popular framework for building LLM applications with a configurable compatible API base and model.OpenAI Compatibleclaude-memA memory extension for Claude Code that can configure an Anthropic-compatible endpoint for its worker.Anthropic CompatibleSillyTavernA power-user frontend for roleplay and long-form chat with multiple APIs and custom model providers.OpenAI CompatibleHammerAIA local-first AI character chat app with a Proxy Mode for compatible model services.OpenAI CompatibleJanitor AIA character creation and immersive chat platform that can use a custom proxy for model requests.OpenAI Compatible

Choose the app you use and follow the setup steps for its API base, key, and model.

  1. Set up Claude CodeAdd TokenHub with an Anthropic-compatible API base, key, and model.
  2. Set up Roo CodeAdd an OpenAI-compatible provider from VS Code settings.
  3. Set up Qwen CodeConfigure a compatible API for terminal coding workflows.
  4. Use a custom model in Kilo CodeChoose a compatible provider for VS Code, JetBrains, or the CLI.
  5. Configure a Janitor AI proxyConnect a compatible model through a custom proxy address.
  6. Connect SillyTavern to an APIAdd an API base, key, and model in connection settings.
  7. Set up OpenHandsAdd an OpenAI-compatible API base, key, and model in LLM settings.

Will this app work with TokenHub?

Check its supported API format, where to add settings, and whether any extra setup is needed.

API format
Check whether the app supports an OpenAI-, Anthropic-, or Gemini-compatible API.
Where to configure it
Look for the API base, API key, and model fields in settings, environment variables, or config files.
What the model needs
Choose a model based on tool use, streaming, context length, and cost.

Browse by what you do

From coding to character chat, explore apps by the way you work.

Most OpenAI-compatible apps use an API base ending in /v1. Anthropic-compatible apps such as Claude Code use the Messages API.

Coding agents

Read code, edit files, run tests, and automate development work from the terminal.

11 appsIDEs & editors

Use models from VS Code, JetBrains, or a standalone code editor.

5 appsPersonal agents

Connect messaging, browsers, and local tools for ongoing personal workflows.

2 appsMulti-model chat

Use a self-hosted or desktop chat client to switch between models in one place.

1 appsDeveloper platforms

Build AI applications with an SDK, gateway, or agent framework.

2 appsRoleplay & characters

Build characters, stories, and long conversations with flexible model access.

3 appsSecurity agents

Run security testing, vulnerability validation, and automated reporting workflows.

1 apps

Questions, answered

Which apps can connect to TokenHub?

This page lists common apps that can be configured with TokenHub through a custom API. Open the relevant setup guide and use its API format and configuration steps to enter the API base, API key, and model name.

What is the difference between OpenAI-compatible and Anthropic-compatible APIs?

They use different request formats and API bases. Cline and Open WebUI usually use an OpenAI-compatible API; Claude Code uses an Anthropic Messages-compatible API. Follow the guide for your app.

What do I need to enter during setup?

Most apps need an API base, a TokenHub API key, and a model name. You may add them in settings, environment variables, or a config file.

How can I confirm that my app is working?

After following the guide, send a simple request with a model available on TokenHub, then test streaming, tool use, or other features you need. Available features can vary by app version and deployment.

What should I do if I get an authentication, model not found, or connection error?

First, confirm that your API base matches the guide, your API key is valid, and the model name is available on TokenHub. Then restart the app and check its logs or error message. If the issue continues, contact support with the app version and error details.

How am I billed for models used through TokenHub?

Model usage is charged according to TokenHub’s current pricing and your actual usage, and may be separate from an app subscription. Check the TokenHub pricing and usage pages for available models, rates, and consumption details.