freellmapi vs router

LLM request routers compared

freellmapi and router both provide a shared endpoint for routing AI requests across configured providers. freellmapi focuses on free provider tiers, custom endpoints, and personal experimentation, while router emphasizes model scoring, routing inspection, and centralized use by applications and agents.

freellmapirouter
LanguageTypeScriptGo
LicenseMITApache-2.0
Stars30.3k5.6k
Forks4.2k159
Last analyzedOct 3, 2026Oct 20, 2026

Key differences

  • freellmapi is written in TypeScript and licensed under MIT; router is written in Go and its v0.2.24-and-later release line is Apache-2.0.
  • freellmapi tracks quotas, rotates keys, and falls back on rate limits or server errors; router uses an in-process cluster scorer to select an enabled model.
  • freellmapi supports optional Ollama-compatible access and common media workflows; router highlights streaming, tool use, and vision across Anthropic, OpenAI, and Gemini formats.
  • freellmapi offers a request playground, routing chains, and usage analytics; router adds route inspection, OTLP trace export, and routing analytics endpoints.
  • freellmapi is positioned for personal experimentation and warns against production use; router describes use by teams, applications, and agent or coding-tool developers.
  • freellmapi can run locally or in Docker and also describes desktop installers; router offers hosted use or self-hosting, with its full self-hosted stack requiring additional setup including Postgres.

Choose freellmapi if you…

  • want to combine free provider tiers and custom OpenAI-compatible endpoints behind one API.
  • need quota tracking, key rotation, and fallback across providers for experiments or development.
  • want a dashboard and client helpers for personal projects, including local-model routing.
Read the freellmapi analysis →

Choose router if you…

  • need centralized model routing for applications, agents, or coding tools.
  • want to inspect routing decisions or export traces for analysis.
  • prefer a Go-based proxy with hosted and self-hosted options and encrypted provider keys at rest.
Read the router analysis →

Compare the numbers side by side, or add more projects

This comparison is generated with AI from the OSRepos analyses of both projects. Always check each project's repository and documentation before choosing.