ENTERPILOT/GoModel
AI gateway written in Go. Lightweight unified OpenAI-compatible API for OpenAI, Anthropic, Gemini, Groq, xAI & Ollama. LiteLLM alternative with observability, guardrails, streaming, costs and usage tracking.
What it solves
GoModel is a high-performance AI gateway designed to simplify the management of multiple LLM providers. It solves the problem of vendor lock-in and operational overhead by providing a single, unified entry point for various AI models, while reducing costs through caching and intelligent routing.
How it works
It acts as a proxy layer between your application and various LLM providers (such as OpenAI, Anthropic, and Google Gemini). It exposes OpenAI-compatible and Anthropic-compatible APIs, allowing developers to switch providers by changing a base URL without rewriting their code. The gateway handles authentication, routing, and monitoring centrally.
Who it’s for
It is built for developers and organizations that use multiple AI models and need a centralized way to track spending, enforce rate limits, and ensure reliability through failover mechanisms.
Highlights
- Unified API: Supports a wide range of providers with OpenAI and Anthropic SDK compatibility.
- Cost Optimization: Includes exact and semantic caching, prompt compression, and spending tracking with hard budget limits.
- Reliability: Features automatic failover to backup providers, retries, and circuit breakers.
- Management Tools: Provides a dashboard for usage analytics, observability, and the ability to create "virtual models" for load balancing.
- Advanced Control: Offers rate limiting, request guardrails, and provider key rotation to bypass per-key limits.