aiproxy

Local proxy for LLM APIs. One endpoint for all providers.

Installation

npm install -g pnkd-aiproxy

Quick Start

# Configure API keys
aiproxy config set openai-key sk-xxx
aiproxy config set anthropic-key sk-ant-xxx

# Start proxy
aiproxy start --port 8080

# Point your app to the proxy
# OpenAI: http://localhost:8080/v1
# Anthropic: http://localhost:8080/anthropic

Commands

FREE Commands

| Command | Description | |---------|-------------| | start | Start proxy server | | stop | Stop proxy server | | config <get\|set> [key] | Manage configuration | | providers | List available providers | | logs | View request logs | | stats | Show statistics | | cost | Show cost breakdown | | limit <set\|clear> [rate] | Set rate limit | | export | Export logs | | license | Manage license |

PRO Commands

| Command | Description | |---------|-------------| | keys <add\|remove\|list> | Manage multiple API keys | | rules | Custom routing rules | | dashboard | Web monitoring dashboard |

Provider Endpoints

OpenAI:    http://localhost:8080/v1
Anthropic: http://localhost:8080/anthropic
Google:    http://localhost:8080/google     [PRO]
Groq:      http://localhost:8080/groq       [PRO]
Ollama:    http://localhost:8080/ollama     [PRO]
Custom:    http://localhost:8080/custom     [PRO]

Options

Start Options

• --port <port> - Proxy port (default: 8080)

Stats/Logs Options

• --last <n> - Show last N entries
• --today - Filter by today
• --month - Filter by this month
• --by-model - Group by model
• --by-day - Group by day
• --json - Output as JSON

FREE vs PRO

| Feature | FREE | PRO | |---------|------|-----| | Requests per day | 100 | Unlimited | | Providers | 2 (OpenAI, Anthropic) | 6 (all) | | Keys per provider | 1 | Unlimited | | Load balancing | - | Yes | | Auto-fallback | - | Yes | | API translation | - | Yes | | Web dashboard | - | Yes | | Custom routing | - | Yes |

PRO Features

Load Balancing

Distribute requests across multiple API keys:

aiproxy keys add openai sk-key1
aiproxy keys add openai sk-key2
aiproxy keys add openai sk-key3

Supports strategies: round-robin, random, least-latency, weighted.

Auto-Fallback

Automatically retry with another provider if one fails:

OpenAI error → Try Anthropic → Try Groq

API Translation

Use OpenAI format with any provider:

# Send OpenAI format to Anthropic
curl http://localhost:8080/anthropic/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{"model": "claude-4.5-opus", "messages": [{"role": "user", "content": "Hi"}]}'

Web Dashboard

Monitor requests, costs, and errors in real-time:

aiproxy dashboard
# Opens http://localhost:8081

PRO License

$18.99 one-time payment

Purchase at: https://pnkd.dev/aiproxy

# Activate license
aiproxy license activate APX-XXXX-XXXX-XXXX-XXXX

# Check status
aiproxy license status

Example Usage

Python with OpenAI SDK

import openai

client = openai.OpenAI(
    api_key="your-key",
    base_url="http://localhost:8080/v1"
)

response = client.chat.completions.create(
    model="gpt-5-turbo",
    messages=[{"role": "user", "content": "Hello"}]
)

JavaScript/TypeScript

import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: 'your-key',
  baseURL: 'http://localhost:8080/v1',
});

const response = await client.chat.completions.create({
  model: 'gpt-5-turbo',
  messages: [{ role: 'user', content: 'Hello' }],
});

cURL

curl http://localhost:8080/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk-xxx" \
  -d '{
    "model": "gpt-5-turbo",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

Cost Tracking

Supported models and pricing (per 1M tokens):

| Model | Input | Output | |-------|-------|--------| | gpt-5 | $20 | $60 | | gpt-5-turbo | $5 | $15 | | gpt-4.5-turbo | $1 | $3 | | o3 | $10 | $40 | | claude-5 | $20 | $80 | | claude-4.5-opus | $15 | $75 | | claude-4.5-sonnet | $3 | $15 | | claude-4.5-haiku | $0.25 | $1.25 | | gemini-2.0-pro | $2.50 | $10 | | llama-4 | $0.50 | $1.50 |

Data Storage

All data stored locally in ~/.aiproxy/:

• config.json - Configuration
• stats.json - Statistics
• logs/ - Request logs (JSONL format)
• license.json - License info

More PRO Tools

• ctxstuff PRO - Pack code for LLMs ($14.99)
• llmcache PRO - Cache LLM responses ($18.99)

Support

• Issues: https://github.com/pnkd-dev/aiproxy/issues
• Website: https://pnkd.dev/aiproxy

License

MIT - pnkd.dev