Skip to content

PlatformsPerplexity

Perplexity Orchestrator API: one API for 45+ frontier and open-weight models with routing and unified billing

Orchestrator API at a glance

Perplexity launched the Orchestrator API on November 17, 2026: one API and one bill for more than 45 frontier and open-weight models from OpenAI, Anthropic, Google and xAI, with provider failover, preview model routing and billing at each model's published rates.

Ask your AI about this story

Ask what changed in Orchestrator API and what to check before using it in your business. Weigh pricing, what it connects to, setup effort and how hard it would be to leave later.

Sources checked Primary source (opens in a new tab)

What changed with Orchestrator API

Perplexity launched the Orchestrator API on November 17, 2026, opening the model layer behind its own products to developers: one API serving more than 45 frontier and open-weight models from OpenAI, Anthropic, Google and xAI, plus open-weight models such as GLM, Kimi and DeepSeek. Perplexity says its production layer routes hundreds of millions of requests across 85+ models and more than a dozen providers.

The Orchestrator API accepts the OpenAI Chat Completions, OpenAI Responses and Anthropic Messages formats, so teams keep their existing OpenAI or Anthropic SDK and change only the API key and base URL. Perplexity tracks each provider's health, capacity and latency, and when a request fails it retries on another deployment of the same model; in one month of production traffic (October 4 to November 2, 2026), Perplexity says Claude and GPT requests through Orchestrator succeeded more often than through any individual provider it measured.

Automatic model routing is in preview: with Routes, teams describe in plain language which requests each model should handle — starting from Lower cost, Balanced or Higher quality profiles — and a decision model matches each request to a route. In Perplexity's evaluation on a 63-task subset of Terminal-Bench 4.1, Claude Fable 5.2 alone solved 38 tasks, while routing cybersecurity, biology and after-refusal requests to GPT-6.1 Sol or GPT-6 Astra solved 41, at a higher estimated cost per task.

Billing runs at each model's published per-token rate with no additional platform fee, under one vendor and one bill; the API Console breaks spend down by model, project, API key and date, with organization model allowlists and spend caps. A second preview feature, Auto effort, adjusts how much reasoning a supported model uses on its next turn under a developer-defined policy.

Why Orchestrator API matters for businesses

Why it matters: Businesses building on AI usually end up with several model providers, each with its own key, contract and invoice. Orchestrator lets a business connect once, switch models as prices and quality change, and add automatic failover and routing without building that plumbing itself — a practical way to cut downtime and keep AI spend visible in one place.

Put it to work: AI services

Get AI News by email

Verified AI stories, each with its official source, on Monday and Thursday mornings. Add breaking news to hear about the biggest ones as soon as we publish them.

Every story links its official source

Free, in English or Spanish

Unsubscribe with one click, any time

What to send

Sources for this story

  1. Official
  2. Official
  3. Official

Last verified: . IA Compa summarizes public announcements for business readers. We are not affiliated with the companies mentioned. Always confirm details, pricing, and availability with the official source.

  • #perplexity
  • #orchestrator-api
  • #model-routing
  • #ai-api
  • #unified-billing
  • #failover
  • #openai
  • #anthropic
  • #google
  • #xai

Want this working in your business?

We turn new AI capabilities into agents, voice receptionists, and automations you can actually use.