Delivery infrastructure for agents. Arch is a models-native proxy server that handles the plumbing work in AI: agent routing & orchestration, guardrails, zero-code logs and traces, unified access to LLMs from OpenAI, Anthropic, Ollama, etc. Build agents faster, and deliver them reliably to production.

A curated list of Awesome-LLM-Ensemble papers for the survey "Harnessing Multiple Large Language Models: A Survey on LLM Ensemble"

High-performance lightweight proxy and load balancer for LLM infrastructure. Intelligent routing, automatic failover and unified model discovery across local and remote inference backends.

RouterArena: An open framework for evaluating LLM routers with standardized datasets, metrics, an automated framework, and a live leaderboard.

Compose, train and test fast LLM routers

NeurIPS 2025 paper "MESS+: Dynamically Learned Inference-Time LLM Routing in Model Zoos with Service Level Guarantees"

Unified interface server for various LLM providers with OpenAI API format

Enrutador inteligente de LLMs para el IndesIAhack 2025 que selecciona el mejor modelo por consulta para minimizar coste y consumo energético manteniendo la calidad.

Hybrid AI routing: LOCAL Ollama + CLOUD GitHub Copilot