X：Aravind Srinivas（Perplexity CEO）

Today we’re open-sourcing Lily, the local inference engine we built for hybrid compute in Perplexity Computer. Lily is specialized for Qwen3.6-35B-A3B on Apple silicon, built so on-device compute doesn’t bottleneck Computer tasks. Read more: https://t.co/OnowTP3ql6

Perplexity Search debuts on the Artificial Analysis Search Index, with all three context size variants taking top positions on the leaderboard The @perplexity_ai Search API comes with three context settings (low, medium, and high) that control how much extracted content each search result carries. We tested all three variants using our standardized methodology: the same model (GPT-5.6 Luna at medium reasoning), running inside Stirrup, our open-source agent harness, with tools for searching and fetching pages from the web. Only the provider behind the search tool changes. Key results: ➤ Perplexity Search (medium) scores 80 on the Artificial Analysis Search Index, ahead of the previous leaders, Parallel (advanced) and Brave Search (LLM context), at 75. The high and low variants score 79 and 77 respectively. Its lead is concentrated in BrowseComp results, with AA-Omniscience and DeepSearchQA scoring comparably to other leading providers ➤ Efficient search payloads: smaller overall search results mean the model reads less per task, so Perplexity has the lowest model inference cost per task of providers we’ve tested so far, ranging from $0.028 to $0.034 across the three variants vs $0.036 for the next lowest provider ➤ Total cost per task is ~$0.091 for the medium and high context variants, at mid-pack latency. For comparison, Parallel (advanced) costs $0.084 per task and Brave (LLM context) costs $0.13 per task

Brain is our self-improving memory system for Perplexity Computer. It compiles sessions, files, and sources into a structured knowledge wiki. New evals build on our initial results, improving correctness by 9.3 points, currentness by 8.0, and recall by 8.9 with 15% fewer tokens. https://t.co/mDMVWt2xzS

New research: Portable Computer is a local-first agent for private and cost-effective work. With an on-device 27B model, our harness scores 82.6% on real knowledge work, beating open-source harnesses Pi and Hermes. Our post-trained PPLX 27B reaches 85.4%. https://t.co/Rb6d47clCI

Today we’re launching Portable Computer on @NVIDIA DGX Spark. Portable Computer is a fully local version of Perplexity Computer, where the entire runtime: orchestrator LLM, subagent LLM, agent harness all run on your local hardware. No cloud dependency. https://t.co/plVWz5PaAw

Computer now works in email. Send, forward, or cc computer@perplexity.com on any thread. Every email task runs as a normal session in Computer, viewable on web and mobile, with the same audit trail as any task in the app. https://t.co/y5UtNtI3Y4

MiniMax Agent now uses @perplexity_ai Search. We benchmarked 3 AI-native search providers across 700+ agent tasks. Perplexity delivered the best combination of answer quality and snippet density: more useful evidence per token, less irrelevant context, and lower end-to-end cost. Compared with Serper, our previous default: ⚡Tool calls per task: 17.8 vs 32.6 (45% reduction) 💰Token usage: 94.6M vs. 162.3M (42% reduction) 🔍Pass rate: +2% increase Total cost: 27% decrease in total cost savings In agent workflows, search is not a one-shot lookup. It is a loop. Better snippets mean better grounding. Better grounding means fewer searches, less context, better answers, and lower cost. One good search can save 14 bad ones. Now shipping in MiniMax Agent. agent.minimax.io/