AI Gateway
Most popularUnified access to 200+ LLMs
Access GPT-4o, Claude 3.5, Gemini 2, Llama 3.3, and 200+ more models through one unified API. Features automatic fallbacks between models, response caching to reduce costs and latency, streaming support, and detailed usage tracking. Write once, access every model.
Quick Start
Install the SDK: npm install @sylphx/sdk
Initialize with your API key
Call sylphx.ai.chat() with your prompt
Handle the response or stream
Documentation
Key Concepts
Chat Completions
Send messages to AI models and receive intelligent responses. Supports streaming for real-time output.
Model Selection
Choose from 200+ models including GPT-4o, Claude 3.5, Gemini, and Llama. Each has different capabilities and pricing.
Fallbacks
Automatically try alternative models if the primary one fails or is rate-limited.
Token Counting
Track input and output tokens for cost management and rate limiting.
SDK Methods
| Method | Description |
|---|---|
sylphx.ai.chat() | Send a chat completion request |
sylphx.ai.embed() | Generate text embeddings |
sylphx.ai.image() | Generate images from text |
sylphx.ai.models() | List available models |
Need more details?
Check out our complete documentation for in-depth guides and tutorials.
View full documentation