KeyCompute

简体中文 | English | 繁體中文 | Español | العربية

Next-generation high-performance AI token compute service platform

Features • Quick Start • Configuration • Project Structure

---

Overview

KeyCompute is a high-performance, extensible, and out-of-the-box AI token compute service platform. It provides enterprise-grade capabilities including unified LLM access, smart routing, metering and billing, multi-level distribution, and observability.

Note: This project is for personal learning only. You must use it in compliance with OpenAI Terms of Use and applicable laws and regulations. Do not use it for illegal purposes. In accordance with the Interim Measures for the Administration of Generative Artificial Intelligence Services, do not provide any unregistered generative AI services to the public in China.

---

Features

Multi-model support

Access all major models through the standard OpenAI API format out of the box:

Provider	Model Families	Status
🟢 OpenAI	GPT-5/GPT-4/GPT-4o/...	✅
🟣 Anthropic	Claude 4/3.7/3.5/...	✅
🔵 Google	Gemini 3/2.5/2.0/...	✅
🔴 DeepSeek	DeepSeek-V4/V3/R1/...	✅
🟠 Zhipu	GLM-5.1/5/4.7/...	✅
🔴 MiniMax	MiniMax-M2.7/M2.5/...	✅
🟤 Ollama	Local models (Llama/Qwen/...)	✅
🟡 vLLM	Self-hosted models	✅

Smart routing

• Two-layer routing engine: model-level routing + account-pool routing
• Load balancing: weighted random allocation across multiple accounts
• Automatic retry on failure: switch channels automatically when a request fails
• Health checks: monitor provider availability in real time

Billing & payments

• Real-time billing: request-level price snapshots with precise post-settlement
• Online top-up: Alipay and WeChat Pay
• Usage analytics: detailed token consumption breakdowns
• Balance management: top-up and spending tracking

Referral distribution

• Referral rewards: earn rewards for inviting new users
• Distribution rules: flexibly configure commission ratios
• Revenue analytics: view referral earnings in real time
• Invite links: generate exclusive invite links with one click

Users & permissions

• Multi-user support: user registration, login, and permission management
• Email verification: signup email verification and password reset
• API key management: create, delete, and view API keys
• Group-based rate limiting: user-level request throttling

Observability

• Prometheus metrics: request volume, latency, and error rate
• Structured logging: JSON logs for easier analysis
• Health check endpoint: /health for service status monitoring

---

Quick Start

Requirements

Component	Version Requirement
Rust	≥ 1.92
Axum	≥ 0.8.0
Dioxus	≥ 0.7.1 (frontend development)
PostgreSQL	≥ 16
Redis	≥ 7 (optional, for distributed rate limiting)
Docker	Latest (container deployment)

Option 1: Docker Compose deployment (recommended)

# Clone the project
git clone https://github.com/your-org/keycompute.git
cd keycompute

# Copy and edit environment variables
cp .env.example .env
# Edit .env and fill in real configuration values

# Start all services
docker compose up -d

# Check service status
docker compose ps

After deployment, visit http://localhost:8080 to get started.

Default account: admin@keycompute.local, password: 12345

Change the default administrator password immediately in production.

Option 2: Local development

# Create the network
docker network create keycompute-internal

# PostgreSQL (using the password from .env)
docker run -d \
  --name keycompute-postgres \
  --network keycompute-internal \
  -e POSTGRES_DB=keycompute \
  -e POSTGRES_USER=keycompute \
  -e POSTGRES_PASSWORD="ObpipdGz00wLxK1u1OupDP4rWVu1NEUpB5QGIiIGbek=" \
  -p 5432:5432 \
  -v keycompute_postgres_data:/var/lib/postgresql/data \
  --restart unless-stopped \
  postgres:16-alpine

# Redis (using the password from .env)
docker run -d \
  --name keycompute-redis \
  --network keycompute-internal \
  -p 6379:6379 \
  -v keycompute_redis_data:/data \
  --restart unless-stopped \
  redis:7-alpine \
  redis-server \
  --requirepass "1VoCAza2HoaOmCafAdM+oxj165CiYpgp2XmD9tTeLN0=" \
  --maxmemory 256mb \
  --maxmemory-policy allkeys-lru

# Install dioxus-cli
curl -sSL http://dioxus.dev/install.sh | sh

# Start the backend service
export KC__DATABASE__URL="postgres://keycompute:ObpipdGz00wLxK1u1OupDP4rWVu1NEUpB5QGIiIGbek=@localhost:5432/keycompute"
export KC__REDIS__URL="redis://:1VoCAza2HoaOmCafAdM+oxj165CiYpgp2XmD9tTeLN0=@localhost:6379"
export KC__AUTH__JWT_SECRET="ea2fe6dd660639d1401c0c4c9fbd71cfe627785ae2359f3b0179efa7c0e24245f966a586295ed598db795da5a942dff7"
export KC__CRYPTO__SECRET_KEY="H8AS+HwrYBp/KSAWRLh9jcLnsV+SIvOtohDPRun+GXA="
export KC__EMAIL__SMTP_HOST="smtp.example.com"
export KC__EMAIL__SMTP_PORT="587"
export KC__EMAIL__SMTP_USERNAME="noreply@example.com"
export KC__EMAIL__SMTP_PASSWORD="your-smtp-password"
export KC__EMAIL__FROM_ADDRESS="noreply@example.com"
export KC__EMAIL__VERIFICATION_BASE_URL="https://api.example.com"
export KC__DEFAULT_ADMIN_EMAIL="admin@keycompute.local"
export KC__DEFAULT_ADMIN_PASSWORD="12345"

cargo run -p keycompute-server --features redis

# Start the frontend development server (in another terminal)
API_BASE_URL=http://localhost:3000 dx serve --package web --platform web --addr 0.0.0.0

---

Project Structure

keycompute/
├── crates/                    # Backend core modules (Rust)
│   ├── keycompute-server/      # Axum HTTP service
│   ├── keycompute-types/       # Shared types
│   ├── keycompute-db/          # Database access layer
│   ├── keycompute-auth/        # Authentication and authorization
│   ├── keycompute-ratelimit/   # Distributed rate limiting
│   ├── keycompute-pricing/     # Pricing engine
│   ├── keycompute-routing/     # Smart routing
│   ├── keycompute-runtime/     # Runtime state
│   ├── keycompute-billing/     # Billing and settlement
│   ├── keycompute-distribution/# Referral distribution
│   ├── keycompute-observability/# Observability
│   ├── keycompute-config/      # Configuration management
│   ├── keycompute-emailserver/ # Email service
│   ├── llm-gateway/            # LLM execution gateway
│   └── llm-provider/           # Provider adapters
│       ├── keycompute-openai/  # OpenAI/Claude/Gemini
│       ├── keycompute-deepseek/# DeepSeek
│       ├── keycompute-ollama/  # Ollama local models
│       └── keycompute-vllm/    # vLLM self-hosted models
├── packages/                   # Frontend (Dioxus 0.7)
│   ├── web/                    # Web admin dashboard
│   ├── ui/                     # Shared UI components
│   └── client-api/             # API client
├── nginx/                      # Nginx configuration
├── Dockerfile.server           # Backend image
├── Dockerfile.web              # Frontend image
└── docker-compose.yml          # Container orchestration

---

Configuration

Environment variables

Primary environment variables:

Variable	Description	Required
KC__DATABASE__URL	PostgreSQL connection string	✅
KC__REDIS__URL	Redis connection string	⚪
KC__AUTH__JWT_SECRET	JWT signing secret	✅
KC__CRYPTO__SECRET_KEY	API key encryption secret	✅
KC__EMAIL__SMTP_HOST	SMTP host	⚪
KC__EMAIL__SMTP_PORT	SMTP port	⚪
KC__EMAIL__SMTP_USERNAME	SMTP username	⚪
KC__EMAIL__SMTP_PASSWORD	SMTP password	⚪
KC__EMAIL__FROM_ADDRESS	Sender email address	⚪
KC__EMAIL__FROM_NAME	Sender display name	⚪
KC__EMAIL__VERIFICATION_BASE_URL	Base URL for email verification and password reset links	⚪
KC__DEFAULT_ADMIN_EMAIL	Default administrator email	⚪
KC__DEFAULT_ADMIN_PASSWORD	Default administrator password	⚪

💡 Tip: Once data has been written to the database, KC__CRYPTO__SECRET_KEY must not be changed, or historical data will no longer be decryptable.

---

API

OpenAI-compatible API

# Chat Completions
curl https://your-domain/v1/chat/completions \
  -H "Authorization: Bearer sk-xxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-4",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

# Example
curl -s http://192.168.100.100:3000/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk-329939d678d24433bc0277311c576481bc23b86ebc724354" \
  -d '{"model":"deepseek-chat","messages":[{"role":"user","content":"hello"}],"stream":true}'

data: {"id":"chatcmpl-7370f2606a6a4f5fa516fe54d9196c9d-kc","object":"chat.completion.chunk","created":1775231430,"model":"deepseek-chat","system_fingerprint":"fp_deepseek","choices":[{"index":0,"delta":{"role":"assistant","content":"Hello! 👋 Nice to meet you!\nHow can I help today?"},"finish_reason":null}]}

# List models
curl https://your-domain/v1/models \
  -H "Authorization: Bearer sk-xxx"

Admin API

Endpoint	Description
POST /api/v1/auth/register	User registration
POST /api/v1/auth/login	User login
GET /api/v1/me	Get current user
GET /api/v1/keys	List my API keys
POST /api/v1/keys	Create an API key
GET /api/v1/usage	Usage statistics
GET /api/v1/billing/records	Billing records
POST /api/v1/payments/orders	Create a payment order

---

Development Guide

# Build
cargo build --workspace --exclude desktop --exclude mobile --verbose

# Run tests
cargo test --lib --workspace --exclude desktop --exclude mobile --verbose
cargo test --package client-api --tests --verbose
cargo test --package integration-tests --tests --verbose

# Code checks
cargo clippy --workspace --exclude desktop --exclude mobile --all-targets --all-features --future-incompat-report -- -D warnings
cargo fmt --all --check

# Enable the Redis backend
cargo build -p keycompute-server --features redis

---

Contributing

We welcome contributions of all kinds. Please read CONTRIBUTING.md to learn how to get involved.

• 🐛 Report bugs
• 💡 Feature requests
• 🔧 Submit code

---

License

This project is open sourced under the MIT License.

---

💖 Thanks for using KeyCompute

If this project helps you, feel free to give it a ⭐️ star.

Quick Start • Report Issues • Latest Releases