AI Gateways
One API, every model.
Browse AI gateways that unify access to hundreds of models across providers. Add routing, fallbacks, caching, and observability to a single integration.
Vercel AI Gateway
Featuredby Vercel
Access models from OpenAI, Anthropic, Google, Meta, xAI, and more through a single interface. Built into the AI TOOLKIT with automatic authentication on Vercel, BYOK support, provider routing, fallbacks, and observability in your Vercel dashboard.
OpenRouter
Featuredby OpenRouter
A unified API gateway with one API key for hundreds of models from Anthropic, Google, Meta, Mistral, and more. Transparent per-token pricing, enterprise-grade infrastructure with automatic failover, and immediate access to newly released models.
Cloudflare AI Gateway
Featuredby Cloudflare
Access models from OpenAI, Anthropic, DeepSeek, Google AI Studio, and more through Cloudflare's AI Gateway. Automatic fallback between models, response caching, retries, rate limiting, and native support for Cloudflare Workers AI bindings.
Portkey
Featuredby Portkey
A production-ready AI gateway with interoperability across 250+ LLMs, full-stack observability and tracing, 50+ guardrails, semantic caching, and conditional request routing with fallbacks, load balancing, and automatic retries.
LangDB
by LangDB
A high-performance enterprise AI gateway built in Rust that governs, secures, and optimizes AI traffic. OpenAI-compatible APIs, access to 250+ models, dynamic request routing, tracing, and cost optimization. Connect to multiple LLMs by changing just two lines of code.
Requesty
by Requesty
A unified LLM gateway with access to over 300 models. 99.99% uptime SLA with intelligent failover and load balancing, prompt caching that cuts costs by up to 80%, prompt injection detection, and built-in real-time observability.
Apertis
by Apertis
A unified AI gateway providing access to 470+ models from OpenAI, Anthropic, Google, and more through a single API key with transparent pay-as-you-go pricing. OpenAI-compatible and enterprise ready with automatic failover.
LlamaGate
by LlamaGate
An OpenAI-compatible API gateway with access to 26+ open-source LLMs (Llama, Qwen, DeepSeek, Mistral) at competitive prices. Includes vision, reasoning, code, and embedding models with pay-as-you-go pricing from $0.02-$0.55 per 1M tokens.
FAQ
Routing, pricing, and lock-in
Language, embedding, and image models from every major lab and provider — routed through a single API. Browse the full catalog on the Models tab.