Continuum-AI-Corp/OrcaRouter-Lite

Self-hosted LLM router with a managed safety net. OpenAI-compatible. BYOK. Single-workspace. Streaming. For more advanced routing choose hosted OrcaRouter

What it solves

OrcaRouter Lite is a self-hosted LLM router that allows users to manage multiple AI providers (like OpenAI, Anthropic, and Google) through a single, OpenAI-compatible API. It eliminates the need to write custom routing logic or manage multiple API keys across different applications, while providing a way to reduce costs by automatically selecting the most affordable model that meets specific technical requirements.

How it works

The system acts as a proxy server. Users provide their own API keys (BYOK) for various providers. When a request is sent with model="auto", the router analyzes the request's needs—such as vision capabilities or JSON mode—and selects the cheapest available model from its catalog of over 100 models that can handle the task. It can also be configured to use a hosted version of OrcaRouter as a fallback safety net for models the user doesn't have keys for.

Who it’s for

It is designed for developers and teams who want to self-host their LLM gateway to maintain control over their keys and analytics, but still want the convenience of a managed routing layer and a unified API for tools like Cursor, Aider, and LangChain.

Highlights

  • Automatic Cost Optimization: The model="auto" feature picks the cheapest capable model based on real-time pricing data.
  • OpenAI Compatibility: Works as a drop-in replacement for any SDK or tool that uses the OpenAI Chat Completions protocol.
  • Cross-Provider Prompt Caching: Reduces costs and latency by caching deterministic requests across different AI providers.
  • Built-in Analytics: Includes a local dashboard to track spending, latency, and potential savings compared to high-end models.
  • Managed Fallback: Ability to integrate a hosted API as an upstream provider to ensure service availability.

Related

  • Project
    mozilla-ai/otari

    An OpenAI-compatible LLM gateway that allows you to manage multiple providers, enforce budgets, and track usage from a single self-hosted endpoint.

  • Project
    ellipticmarketing/modelrelay

    An OpenAI-compatible local router that benchmarks and automatically forwards requests to the best available free AI coding models across multiple providers.

  • Project
    diegosouzapw/OmniRoute

    OmniRoute is an open‑source, self‑hosted router that aggregates hundreds of LLM providers (including 90+ free tiers) behind a single OpenAI‑compatible endpoint. It auto‑selects the best backend, falls back on failures, compresses prompts to save tokens, and shows live free‑tier usage. Install with `npm i -g omniroute`, run locally, and call the `auto` model to get free, resilient AI responses without managing multiple keys.

  • Project
    ulab-uiuc/LLMRouter

    An open-source library for intelligent LLM routing that dynamically selects the optimal model for each query to balance cost and performance.

  • Project
    BlockRunAI/ClawRouter

    An open-source smart LLM router built for autonomous agents, using wallet signatures for authentication and USDC micropayments to eliminate the need for accounts and credit cards.