Tingly Box

Quick Start • Features • Usage • Documentation • Issues

Tingly Box decides which model to call, when to compress context, and how to route requests for maximum efficiency, offering secure, reliable, and customizable functional extensions.

Key Features

• Unified API – One mixin endpoint to rule them all, use what you like - OpenAI / Anthropic / Google
• Smart Routing, Not Just Load Balancing – Intelligently route requests across models and tokens based on cost, speed, or custom policies, not simple load balancing
• Smart Context Compression – (Coming soon) Automatically distill context to its essential parts: sharper relevance, lower cost, and faster responses
• Auto API Translation – Seamlessly bridge OpenAI, Anthropic, Google, and other API dialects—no code changes needed
• Blazing Fast – Adds typically < 1ms of overhead—so you get flexibility without latency tax
• Flexible Auth – Support for both API keys and OAuth (e.g., Claude.ai), so you can use your existing quotas anywhere
• Visual Control Panel – Intuitive UI to manage providers, routes, aliases, and models at a glance
• Client Side Usage Stats - Track token consumption, latency, cost estimates, and model selection per request—directly from your client

Quick Start

Install

From npm (recommended)

# Install and run (auto service migration without any args)
npx tingly-box@latest

if any trouble, please check tingly-box process and port 12580 and confirm to kill them.

From source code

Requires: Go 1.21+, Node.js 18+, pnpm, task, openapi-generator-cli

# Install dependencies
# - Go: https://go.dev/doc/install
# - Node.js: https://nodejs.org/
# - pnpm: `npm install -g pnpm`
# - task: https://taskfile.dev/installation/, or `go install github.com/go-task/task/v3/cmd/task@latest`
# - openapi-generator-cli: `npm install @openapitools/openapi-generator-cli -g`

git submodule update --init --recursive

# Build with frontend
task build

# Build GUI binary via wails3
task wails:build

From Docker (Github)

mkdir tingly-data
docker run -d \
  --name tingly-box \
  -p 12580:12580 \
  -v `pwd`/tingly-data:/home/tingly/.tingly-box \
  ghcr.io/tingly-dev/tingly-box

Use with IDE, CLI, SDK and any AI application

Tool Integration

• Claude Code
• OpenCode
• Xcode
• Gemini
• ……

Any application is ready to use.

OpenAI SDK

from openai import OpenAI

client = OpenAI(
    api_key="your-tingly-model-token",
    base_url="http://localhost:12580/tingly/openai/v1"
)

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

Anthropic SDK

from anthropic import Anthropic

client = Anthropic(
    api_key="your-tingly-model-token",
    base_url="http://localhost:12580/tingly/anthropic"
)

response = client.messages.create(
    model="tingly",
    max_tokens=1024,
    messages=[
        {"role": "user", "content": "Hello!"}
    ]
)
print(response)

Tingly Box proxies requests transparently for SDKs and CLI tools.

Using OAuth Providers

You can also add OAuth providers (like Claude Code) and use your existing quota in any OpenAI-compatible tool:

# 1. Add Claude Code via OAuth in Web UI (http://localhost:12580)
# 2. Configure your tool with Tingly Box endpoint

Requests route through your OAuth-authorized provider, using your existing Claude Code quota instead of requiring a separate API key.

This works with any tool that supports OpenAI-compatible endpoints: Cherry Studio, VS Code extensions, or custom AI agents.

Web Management UI

npx tingly-box@latest

Documentation

User Manual – Installation, configuration, and operational guide

Philosophy

• One endpoint, many providers – Consolidates multiple providers behind a single API with minimal configuration.
• Seamless integration – Works with SDKs and CLI tools with minimal setup.

How to Contribute

We welcome contributions! Follow these steps, inspired by popular open-source repositories:

1. Fork the repository – Click the “Fork” button on GitHub.

2. Clone your fork

   git clone https://github.com/your-username/tingly-box.git
   cd tingly-box
   

3. Create a new branch

   git checkout -b feature/my-new-feature
   

4. Make your changes – Follow existing code style and add tests if applicable.

5. Run tests

   task test
   

6. Commit your changes

   git commit -m "Add concise description of your change"
   

7. Push your branch

   git push origin feature/my-new-feature
   

8. Open a Pull Request – Go to the GitHub repository and open a PR against main.

Support

Telegram	Wechat
https://t.me/+V1sqeajw1pYwMzU1	http://chv.ckcoa5.cn/t/OSFb

Contributors

Mozilla Public License Version 2.0 · © Tingly Dev