Beta
Discord Download

Requesty vs AutoRouter

Requesty is a hosted LLM gateway: one OpenAI-compatible API key for 600+ models, with routing policies, spend limits and analytics. AutoRouter is a Mac app that adds every leading model to Codex, Claude Code and other tools, uses the plans you already pay for, and picks a model for each turn. This page covers how Requesty works, then how the two compare.

AutoRouter A Mac app that picks a model for each turn. $0/mo on your plans Download Requesty A hosted gateway with one key for 600+ models. 5%on model cost
Price $0 on your plans, Pro $50 a month, Enterprise by seat Free for free models, pay as you go, Enterprise custom
Fee on tokens Provider’s price plus 10% on paid models Provider’s price plus 5%
Runs as A Mac app, macOS 14 or later A hosted API, with an EU region
Agents Codex, Claude Code, Cursor, Claude Desktop, T3 and Terminal Claude Code, Codex, Cursor, Cline, Kilo Code, Pi, OpenCode and more
Setup Install, sign in, click Enable Create an API key, then point each agent at Requesty
Your plans Claude, ChatGPT, Grok and Gemini plans, used first Requesty credits, or your own provider keys
Model choice Picked for each turn, after your rules The model you name, or a policy you build in its console
Teams MDM rollout, allowed models, budgets, SSO, SCIM Groups, approved models, spend limits, SSO, audit logs
Requests go Through AutoRouter’s servers to each provider Through Requesty’s gateway to each provider
  • Uses the plans you already pay for
  • A model picked for each turn, with no policy to build
  • No API keys to create or paste
  • Rollout to every Mac in a team through MDM

Sources checked November 16, 2026: Requesty docs, Claude Code guide, Codex guide, Auto Routing, Requesty pricing and AutoRouter pricing. AutoRouter wrote this page. See every comparison.

How Requesty works

What the gateway is, how it connects to coding agents, and how its routing policies pick a model.

One key for 600+ models

Requesty is an OpenAI-compatible gateway at router.requesty.ai for 600+ models from 30+ providers. You use a Requesty API key in place of each provider’s key. It adds fallbacks, load balancing, latency routing, caching, spend limits per key or project, and analytics on spend, latency and errors. An EU endpoint keeps processing and storage in Frankfurt (docs).

A bare model name such as claude-sonnet-5-5 is a managed policy: Requesty picks the best provider for that model on each request and fails over when one degrades. You can also bring your own provider keys and use them in your policies.

How setup works

Sign up, create an API key, and point your tools at Requesty. For Claude Code, set ANTHROPIC_BASE_URL, ANTHROPIC_AUTH_TOKEN and ANTHROPIC_MODEL in settings.json. For Codex, store the key in auth.json and add a Requesty provider to config.toml. The Requesty CLI writes either config for you and backs up the old one. An optional wrapper tags Claude Code requests with your branch, repo and user.

Auto Routing

Requesty’s Auto Routing lets you describe kinds of requests, attach a model to each, and pick a decision model. It reads the first user message of each conversation, picks a route, and keeps that model for the rest of the conversation (Auto Routing).

What it does well

  • One key reaches 600+ models, so you skip separate provider accounts.
  • Its 5% fee is low, with no per-seat price or minimum spend.
  • It works from any system, for coding agents and your own apps alike.
  • Managed policies and fallback chains keep requests going when a provider degrades.
  • Teams get groups, approved models, spend limits, SSO, audit logs and guardrails for PII and secrets.
  • EU data residency is on every plan.

Where it gets harder

  • It runs on Requesty credits or your own provider keys. Its Claude Code and Codex guides don’t use Claude or ChatGPT plans.
  • Auto Routing is yours to design: the routes, the models and the decision model.
  • It picks once per conversation, so a session that turns harder keeps the model chosen at the start.
  • Each developer edits their agents’ config files, or runs the CLI, on each machine.
  • The Claude Code guide says to set the model with ANTHROPIC_MODEL, because Claude might not accept /model.

How AutoRouter differs

AutoRouter is turnkey on a Mac: no API keys, your plans used first, and a model picked for every turn.

The plans you already pay for

You connect Claude, ChatGPT, Grok and Gemini plans, and they go to the work that needs them, so they last longer. You can use them in any harness, such as a Claude plan in Codex. When one account reaches its limit, AutoRouter moves to the one with the most room left before it spends any credit. A subscription isn’t required: Pro gives every model, with $20 of included usage each month.

A model for each turn

Your plain-English rules go first, such as “Use Claude Opus 5.5 for UI work.” Otherwise AutoRouter works out how hard each turn is and only considers models that clear the quality bar for it. A task that continues keeps its model, or moves up. Every turn shows its model, what it cost and why.

Rolled out to a team’s Macs

Enterprise rolls AutoRouter out to every Mac through Jamf or another MDM, with nothing for anyone to set up. Admins get allowed models, team budgets and zero data retention in one place, spend by team, project or ticket, and SSO, SCIM and an audit log.

What it costs and what it can’t do

Routing across your plans is free. Pro is $50 a month, and paid models bill at the provider’s price plus 10%, twice Requesty’s fee. AutoRouter runs only on macOS 14 or later and doesn’t serve your own apps’ API traffic. It sometimes misjudges how hard a task is. See privacy for what is kept.

Which one to choose

Choose Requesty if you want one API key for apps and agents on any system, pay as you go with a low fee, or need EU data residency.

Choose AutoRouter if you work on a Mac, already pay for Claude or ChatGPT plans, and want models picked for you with no keys or config files. Teams can see Enterprise.

Setup, side by side

The steps each one’s own docs give, as of November 16, 2026.

Requesty

  1. Sign up Create an account at app.requesty.ai and add credits.
  2. Create a key Make an API key on the API Keys page.
  3. Connect your agent Run the Requesty CLI, or set the base URL, key and model in each agent’s config.
  4. Add policies Optionally, build fallback, load balancing or Auto policies in the console.

AutoRouter

  1. Download AutoRouter for macOS, signed and notarized by Apple.
  2. Sign in Sign in or create a free account, then sign in to your plans.
  3. Enable your apps Allow Accessibility and click Enable for each app.
  4. Pick AutoRouter In a new session, choose AutoRouter from Codex’s model menu, or with /model in Claude Code.

Questions

What people ask about Requesty.

What is Requesty?

A hosted, OpenAI-compatible LLM gateway. One Requesty API key reaches 600+ models from 30+ providers, with routing policies, caching, spend limits and analytics.

How much does Requesty cost?

Free models are free up to 200 requests a day. Pay as you go adds 5% to the model’s price, so a $10 model costs $10.50. Enterprise pricing is custom.

Does Requesty work with Claude Code and Codex?

Yes. Its guides set Claude Code’s base URL, token and model in settings.json, and add a Requesty provider to Codex’s config.toml. The Requesty CLI can write both.

Does Requesty pick a model automatically?

With an Auto policy, a decision model you choose reads the first message of each conversation and picks one of your routes. AutoRouter picks a model for each turn without a policy to build.

Can Requesty use my Claude or ChatGPT plan?

Its Claude Code and Codex guides use a Requesty API key, with credits or your own provider keys. AutoRouter uses your Claude, ChatGPT, Grok and Gemini plans first.

How is AutoRouter different from Requesty?

Requesty is a hosted API you configure each tool to call. AutoRouter is a Mac app that needs no API keys, uses your plans first, includes models, picks one for each turn, and rolls out to a team’s Macs through MDM.

How do I move over from Requesty?

Install AutoRouter, sign in to your plans and click Enable for each app. Enabling adds AutoRouter to the app’s own settings after saving a copy of the file. Rewrite any routes you rely on as plain-English rules.

What does AutoRouter cost?

Routing across your ChatGPT and Claude plans is free. Pro is $50 a month with $20 of usage, and paid models bill at the provider’s price plus 10%. See pricing, or estimate a month with and without AutoRouter in the savings calculator.