Requesty
A single API endpoint that talks to 600+ AI models from 30+ providers, so switching from one AI model to another — or spreading requests across several for cost or reliability — doesn't mean rewriting your app's code.
🔗 Visit Requesty
Description
Every AI provider (OpenAI, Anthropic, Google and dozens more) has its own API format, so an app that wants to use several models — or switch when a cheaper or better one comes along — normally has to write and maintain separate integration code for each. Requesty removes that friction: it's a single, OpenAI-compatible endpoint that routes a request to whichever of 600+ models across 30+ providers you specify, so switching providers becomes a config change instead of a rewrite.
Requesty is a managed, Rust-built LLM gateway routing requests across 600+ models from 30+ providers including OpenAI, Anthropic, Google, DeepSeek and Mistral through one OpenAI-compatible API. It offers intelligent routing, prompt caching claimed to cut token costs up to 90%, and enterprise governance features (SSO, RBAC, audit logs, model whitelisting). Pricing is straightforward: a Free tier (200 requests/day, free models only), Pay-As-You-Go (a flat 5% markup on underlying model costs, no subscription or seat fees, $10 free credit on signup), and custom Enterprise pricing with SOC2/GDPR/HIPAA compliance and EU data residency.
💬 Our review
The short version: Requesty's core value is straightforward and genuinely useful — one integration instead of 30+ separate provider integrations — and its flat 5% markup pricing model is easier to reason about than gateways with more convoluted fee structures.
The honest gap is competitive crowding: this is a busy category with OpenRouter and Portkey both already established (and both already covered elsewhere in this catalog), so Requesty's differentiation comes down to specifics — its claimed 90% cost reduction via prompt caching and its enterprise governance features (SSO, RBAC, audit logs) aimed at larger teams that need compliance guardrails around multi-provider LLM usage. Its usage stats (70,000+ developers, 90B+ tokens/day) are self-reported without independent verification, so treat them as directional. For teams that specifically need enterprise governance controls around multi-model routing — audit logs, model whitelisting, EU data residency — Requesty's enterprise tier is a reasonable pick; for a simpler, more community-driven routing layer, OpenRouter's broader ecosystem and marketplace-style model listing may already do the job.
💰 Pricing
📊 Global score
🤖 AI-enriched data
Free : 200 requêtes/jour, modèles gratuits uniquement. Pay-As-You-Go : markup fixe de 5% sur le coût des modèles, pas d'abonnement, $10 de crédit gratuit à l'inscription. Enterprise : sur devis (SSO, RBAC, audit logs, résidence des données UE).
Pros
Un seul endpoint compatible OpenAI pour 600+ modèles sur 30+ fournisseurs
Markup fixe de 5%, facile à raisonner par rapport à d'autres structures de frais
Gouvernance enterprise (SSO, RBAC, audit logs, résidence UE) pour les grandes équipes
Cons
Catégorie très concurrentielle (OpenRouter, Portkey déjà bien établis)
Statistiques d'usage auto-rapportées, non vérifiées indépendamment
Pas de trace GitHub publique pour juger de la traction communautaire