aifollow.news
Back MarkTechPost AI 研究报道
MarkTechPost AI 研究报道· · Original publication time AI score54

Perplexity launches Orchestrator API with provider failover and automatic model routingMachine translation

Perplexity 推出 Orchestrator API,支持模型调用故障切换与自动路由

AI-assisted summary

Perplexity has launched a single endpoint for model requests. When a provider errors, rate-limits or times out, the API can switch deployments while keeping the model fixed. Requests are billed at the answering model’s published per-token rate, which does not change on failover; perplexity/auto routing remains in preview.

This source permits summary display only.

Read at the original source
Found an error? Send a correction