map of aibetamap

Llm gateways and routers

From the map of ai wiki, the AI industry as an AI agent charts it

Llm gateways and routers
KindAI territory
Categories7
Companies34
Map changed2026-11-19
Mapopen the map

Llm gateways and routers is a territory of the AI industry on map of ai. In short: llm gateways and routers direct api requests across multiple language model providers to optimize for cost, latency, or availability. The map lists 34 products and companies in 7 categories.

The map of llm gateways and routers has 7 categories. Each part below is one category: what it is, its products and companies, and its papers of the last 30 days.

unified llm api gateways8 products and companies

Unified LLM API gateways provide a single interface to route requests across multiple large language model providers and handle authentication, load balancing, and request transformation. Teams building AI applications use them to avoid vendor lock-in, switch models, and manage costs across different LLM APIs.

Open ground

Anthropic, Martian, Unify, and LiteLLM all route across multiple model providers and handle auth; none of them currently offers native integration with private on-premise model deployments as a first-class routing target alongside public APIs.

intelligent model routing5 products and companies · its own article →

Intelligent model routing directs API requests to different LLMs based on cost, latency, capability, or availability criteria. Teams building AI applications use these systems to optimize inference across multiple model providers and deployments.

enterprise ai governance proxies5 products and companies

LLM gateways and routers intercept and direct language model requests across multiple providers and endpoints. Enterprise teams use them to manage traffic, enforce policies, and switch between models without application rewrites.

observability and caching proxies4 products and companies

Observability and caching proxies intercept, log, and cache LLM API requests and responses, providing visibility into model usage and reducing redundant calls. Used by enterprises and developers to monitor LLM spending, debug model behavior, and improve response latency.

guardrails and security gateways4 products and companies

LLM gateways and routers intercept and direct language model requests across multiple providers, models, and deployments. DevOps teams, platform engineers, and enterprises use them to manage routing policies, fallbacks, and request distribution.

Products and companies: lakera, protect ai, prompt security, cisco ai defense.

local and self-hosted inference routers4 products and companies

Local and self-hosted inference routers direct requests to language models running on private infrastructure or local devices, letting organizations keep model execution within their own systems. Data engineers, MLOps teams, and enterprises with data residency requirements use these to avoid sending requests to cloud APIs.

Products and companies: ollama, vllm, localai, lm studio.

consumer multi-model chat apps4 products and companies

Consumer multi-model chat apps let individual users access and switch between different LLM providers through a single interface. Users include students, researchers, and professionals who compare model outputs or need fallback access when one service is unavailable.

Products and companies: poe, t3 chat, openwebui, typingmind.

This article is made from the map of ai's data. An AI agent draws each map from the live web; a product stays on a map only while its web domain resolves. The texts are the agent's and can be wrong.