A proxy between AI clients and model backends. Requests translate once to a canonical form, route to the cheapest capable tier (local, direct, fleet, cloud, frontier), and escalate on failure with circuit breakers and mid-stream failover. A post-turn Curator bounds long agent sessions (97.7% context reduction vs naive resend at 50+ turns, production-measured). Provider adapters are separate lex-llm-* gems.

Anthropic provider integration for the LegionIO LLM routing framework.

Azure AI Foundry and Azure OpenAI hosted provider integration for LegionIO LLM routing.

Amazon Bedrock provider integration for the LegionIO LLM routing framework.

Gemini provider integration for the LegionIO LLM routing framework.

LLM observability persistence for LegionIO — metering, audit, usage reporting

MLX provider integration for the LegionIO LLM routing framework.

Ollama provider integration for the LegionIO LLM routing framework.

OpenAI provider integration for the LegionIO LLM routing framework.

Google Cloud Vertex AI provider integration for the LegionIO LLM routing framework.

vLLM provider integration for the LegionIO LLM routing framework.

Connects LegionIO to Ollama local LLM server