Requesty
Hosted AI gateway routing to 600+ models through one endpoint.
About
Requesty is a hosted AI gateway that routes your application to 600+ models through a single OpenAI-compatible endpoint. Change one base URL and you get intelligent routing, automatic fallbacks, prompt caching, guardrails, and real-time cost analytics, with the option to bring your own provider keys.
Pricing is a flat 5% markup on model spend with no seats or feature tiers, and the free tier covers 200 requests a day. The tradeoff is routing lock-in: your fallback and caching logic lives on someone else's infrastructure.
What it can do
- Single endpoint for 600+ models
- Smart routing and automatic failover
- Prompt caching and cost optimization
- Guardrails and per-user spend limits
- Bring-your-own provider keys
- Native OAuth for coding tools
Setup
- Create an account at requesty.ai.
- Generate an API key from the dashboard.
- Set your client base URL to https://router.requesty.ai/v1.
- Pick models with provider/model ids like openai/gpt-4o-mini.
Prerequisites
A Requesty account (free tier needs no credit card).
Use cases
Works with
Trust score — 6/10
Scored 6/10 (metrics tier, metrics-v1). Evidence: 600+ models via one endpoint; 5% markup, no seat pricing; free tier: 200 requests/day; closed SaaS — routing lock-in risk. License: Proprietary SaaS.
How the trust score is built
Full methodology →- Adoption · stars/downloads, log-scaled vs peers
- 6.0/10
- Maintenance · recency, releases, issue load
- 7.0/10
- Openness · license + docs
- 3.0/10
- Manual review · sanity read + red-flag check
- 6.0/10
- · 600+ models via one endpoint
- · 5% markup, no seat pricing
- · free tier: 200 requests/day
- · closed SaaS — routing lock-in risk
Sources
Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.