{"slug": "perplexity-launches-orchestrator-api-for-flexible-model-selection", "title": "Perplexity launches Orchestrator API for flexible model selection", "summary": "Perplexity launched its Orchestrator API in private preview on November 9, 2026, a single gateway at https://api.perplexity.ai/orchestrator/v1 that routes requests across Perplexity-hosted open-weight models from Moonshot AI, Z.AI and DeepSeek under one API key and one pay-as-you-go bill. The API accepts both OpenAI and Anthropic request formats regardless of which lab built the model, and adds automatic request routing, health monitoring and failover across deployments, with rejected requests such as rate-limit errors carrying no charge. The launch follows Perplexity's Agent API earlier in 2026 and comes as the company reported an annual recurring revenue run rate exceeding $1 billion as of mid-November 2026, with leadership floating a possible IPO by 2027.", "body_md": "# Perplexity launches Orchestrator API for flexible model selection\n\nPerplexity AI logo, trademark of Perplexity AI, Inc.; logo file public domain on Wikimedia Commons (recolored to brand…\n\nThe private preview routes requests across hosted open-weight models from Moonshot AI, Z.AI and DeepSeek under one key and one bill\n\n[Perplexity](https://cryptobriefing.com/markets/perplexity/) has launched its Orchestrator API, a single gateway that picks models and providers for a given task and rolls the costs into one bill.\n\nThe service went live in private preview on November 9, 2026. It is aimed squarely at developers who are tired of juggling a different integration for every model they want to try.\n\n## What the Orchestrator API actually does\n\nThe API gives developers unified access to a catalog of open-weight models that Perplexity hosts itself. Open-weight means the underlying model parameters are publicly released, so a company like Perplexity can run them on its own infrastructure.\n\nEverything runs through a single base URL: **https://api.perplexity.ai/orchestrator/v1**. Developers authenticate with their existing Perplexity API key, so there is no new credential to manage.\n\nThe more interesting part is compatibility. The API accepts request formats from both [OpenAI](https://cryptobriefing.com/markets/openai/) and [Anthropic](https://cryptobriefing.com/markets/anthropic/), the two schemas most developers already have baked into their code.\n\nPerplexity says developers can use any supported schema regardless of which lab originally built the model. That decoupling is the whole pitch: the format you write in no longer dictates the model you can call.\n\n## The model lineup and the reliability layer\n\nThe initial catalog leans heavily on prominent open-weight releases. Listed models include:\n\n### AI, tech, and the markets they move—in one daily briefing.\n\nDaily. Free. Join 34,000+ readers across crypto, finance, and policy.\n\n- **perplexity/kimi-k3** , from Moonshot AI\n- **perplexity/glm-5.3** and**perplexity/glm-5.3-flash** , from Z.AI\n- **perplexity/deepseek-v4-pro-0813** , from[DeepSeek](https://cryptobriefing.com/markets/deepseek/)\n\nDevelopers can pull the full list programmatically through a **GET /orchestrator/v1/models** call. Pricing is published per million tokens for each model.\n\nEach model ID comes with automatic request routing, health monitoring and failover across all deployments. If one deployment stumbles, traffic gets shifted elsewhere rather than returning an error to the user.\n\n## How billing works\n\nThe billing model is pay-as-you-go, based on token usage at the published per-token rates. Tokens are the small chunks of text that models read and write, so the bill scales with how much work the model actually does.\n\nRejected requests, such as those that hit rate-limit errors, carry no charge.\n\nAccess is not open to everyone yet. Developers who want in during the private preview need to email [\\[email protected\\]](https://cryptobriefing.com/cdn-cgi/l/email-protection).\n\n## Background: Perplexity’s busy 2026\n\nThe Orchestrator API is the latest in a string of developer-focused releases from Perplexity this year. Earlier in 2026, the company introduced its Agent API, and the new launch builds on that push into enterprise tooling.\n\nThe timing also lines up with other November 2026 initiatives, including Multiagent effort levels and Frontier Research projects.\n\nThe business numbers suggest the strategy is landing. Perplexity reported an annual recurring revenue run rate exceeding $1 billion as of mid-November 2026, a figure described as significantly higher than the prior year.\n\nCompany leadership has also floated the possibility of accelerating toward an IPO by 2027 as revenue climbs.\n\n**Disclosure:** This article was edited by Diego Almada Lopez. For more information on how we create and review content, see our [Editorial Policy](https://cryptobriefing.com/editorial-policy/).", "url": "https://wpnews.pro/news/perplexity-launches-orchestrator-api-for-flexible-model-selection", "canonical_source": "https://cryptobriefing.com/perplexity-launches-orchestrator-api/", "published_at": "2026-11-17 20:18:20+00:00", "updated_at": "2026-11-17 21:19:59.939177+00:00", "lang": "en", "topics": ["ai-products", "ai-tools", "large-language-models", "developer-tools", "ai-infrastructure"], "entities": ["Perplexity", "Orchestrator API", "Moonshot AI", "Z.AI", "DeepSeek", "OpenAI", "Anthropic", "Agent API"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/perplexity-launches-orchestrator-api-for-flexible-model-selection", "markdown": "https://wpnews.pro/news/perplexity-launches-orchestrator-api-for-flexible-model-selection.md", "text": "https://wpnews.pro/news/perplexity-launches-orchestrator-api-for-flexible-model-selection.txt", "jsonld": "https://wpnews.pro/news/perplexity-launches-orchestrator-api-for-flexible-model-selection.jsonld"}}