Perplexity Launches Orchestrator API for Failover and Model Routing
Perplexity Launches Orchestrator API for Failover and Model Routing
Perplexity released the Perplexity Orchestrator API, a single endpoint serving frontier models from OpenAI, Anthropic, Google and xAI plus open-weight models such as GLM, Kimi and DeepSeek, with automatic same-model failover across providers and billing through one account. A preview perplexity/auto router selects among 1 to 5 routes from plain-language descriptions, and Perplexity reports that adding routes to GPT-6.1 Sol or GPT-6 Astra raised Terminal-Bench 4.1 results from 38/63 to 41/63 on a 63-task subset. Requests are billed at the answering model's published per-token rate with no additional platform fee on credits, and access is tiered by cumulative credit purchases.
Source: MarkTechPost · marktechpost.com