Answer engine and agentic platform built on RAG over a live web index with inline source citations. Runs a model-agnostic inference stack on NVIDIA H100s using Triton and TensorRT, orchestrating frontier models (Opus, Gemini, Grok, GPT) per task type. Expanding into autonomous multi-step workflows via Perplexity Computer and the Comet AI-native browser.
Perplexity has opened up the model routing layer it uses in production. One API key now reaches 45+ open and proprietary models, with automatic provider failover, one bill, and an optional router that picks a model for each request.
Perplexity's new Multiagent effort level in Computer uses a lead agent to coordinate a mixed team of frontier models, beating single-model runs on Terminal-Bench 4.1 and WANDR.
Perplexity extends its persistent agent product to the Mac desktop, letting Max subscribers delegate long-running jobs across engineering, marketing, QA, and legal.
Perplexity's follow-up red-team study shows that Capture the Flag framing misdirects agents away from the actual security boundary, and cyber-tuned models don't always win.
Perplexity's new Teammates turn its Computer agent into always-on digital coworkers that monitor work, collaborate in Slack, and run in isolated desktops.
Perplexity's agent now runs the model, planner, and tool execution entirely on Apple silicon, keeping private files and credits off the cloud entirely.
Perplexity released SPACE under Apache 2.0, the Firecracker-based microVM sandbox that isolates every agent task in Computer with external credential and egress control.
Perplexity's new 27B and 4B decision models return calibrated probabilities instead of text, answer multiple questions in one pass, and cost two cents per million tokens.
Perplexity's new late-interaction embedding family keeps 128-dim vectors per token, searches rendered pages without OCR, and shares one space across sizes.
Perplexity's agentic Computer tool now renders interactive charts inline, using TradingView's Lightweight Charts for candlesticks, volume, and moving averages.
Perplexity's new 9B contextual embedding model distills relevance from a compression teacher, beating Voyage on turbopuffer's private context-bench with 8x less storage.
Perplexity's Computer agent gains scheduled and event-triggered automations that carry memory across runs, replacing one-shot tasks with persistent background workers.
Perplexity's post-training method teaches its Computer agent to correct mistakes using hint-guided self-distillation, cutting live tool-call failures by 21.2%.
Perplexity Computer now generates finished video clips inline via MiniMax H3 and ByteDance Seedance 2.5, bundled with copy and creative for Pro and Max users.
Perplexity Computer now lets users pick effort levels from Light to Ultra, and the orchestrator picks the right model and reasoning depth automatically.
Perplexity released Q2D-Web, a large-scale benchmark with 190M documents and 70K agent-reformulated queries for evaluating retrieval in agentic RAG systems.
Perplexity's Brain turns agent memory into a linked Markdown wiki that background agents refine, boosting correctness 9.3 points with 15% fewer tokens.
Perplexity's new Portable Computer runs the full agent stack locally on NVIDIA DGX Spark, with cloud escalation gated by user approval and zero per-token cost for on-device work.
Perplexity brings Moonshot AI's record-breaking 2.8T-parameter open-weight model to its search and agentic platform, with a US-only hosting pledge to sidestep data sovereignty concerns.
Perplexity's Personal Computer agent is now live on Windows, bringing multi-model AI orchestration to local files, Microsoft 365, and 400+ connected apps
Perplexity swaps in xAI's freshly-launched Grok 4.5 as the orchestrator brain of Computer, beating every rival configuration on WANDR at roughly half the cost of Claude Opus 4.8
Perplexity's Deep Research is now a native skill inside Computer, powered by a new Search as Code architecture that outperforms every major competitor on research benchmarks