Perplexity introduces multiagent orchestration with model routing
Perplexity has launched a multi-agent orchestration capability allowing users to combine distinct AI models, including those from OpenAI, Anthropic, and Google, into parallel agent streams. This release aims to increase task complexity handling and accuracy by dynamically routing subtasks to the most suitable model.
This shift marks Perplexity's evolution from a search interface to an agentic platform that abstracts model selection. By effectively turning the LLM provider landscape into a commodity component of a larger workflow, they are positioning their product as an essential orchestration layer.
- –Dynamic routing allows for optimizing both latency and performance based on the specific subtask.
- –Parallel agent streams enable faster execution of complex, multi-part user requests.
- –This approach challenges the "one model to rule them all" paradigm, favoring a "best tool for the job" strategy.
DISCOVERED
2h ago
2026-11-17
PUBLISHED
2h ago
2026-11-17
RELEVANCE
AUTHOR
AravSrinivas