Changelog - Perplexity

2026-07-14 · Source: perplexity.ai via Google News · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Data Science & Analytics, Software Development & Engineering · Depth: Intermediate, medium

Summary

Perplexity has released a significant update to its "Computer" product, introducing several new features and enhancements. Key additions include "Brain," a self-improving memory system that builds a private context graph, improving answer correctness by 25%, recall by 16%, and cutting costs by 13% on tasks with prior context. Users can now access Opus 4.8 in a fast mode for quicker complex task responses and utilize Claude Fable 5 as an orchestrator for intricate research synthesis and multi-source analysis. The platform also enables mid-task switching of orchestrator models, allowing dynamic adaptation to workflow stages. Other updates include direct website publishing to pplx.app or custom domains, streamlined homepage navigation, and detailed usage analytics broken down by model. The Mac app gained an account switcher, and Computer now offers private company financial research sourced from Forge Global.

Key takeaway

For AI Engineers and Data Scientists managing complex projects, these Perplexity Computer updates offer significant workflow enhancements. You can utilize "Brain" to automatically retain context from past tasks, reducing rework and improving accuracy. Dynamically switch orchestrator models like Claude Fable 5 mid-task to optimize for cost or reasoning complexity, ensuring efficient resource use. Additionally, the new website publishing and private company research features expand your project delivery and data analysis capabilities directly within the platform.

Key insights

Perplexity's Computer enhances AI agent capabilities with self-improving memory, advanced model orchestration, and integrated web publishing.

Principles

• Contextual memory boosts AI performance.
• Orchestrator models optimize complex workflows.
• Dynamic model selection improves efficiency.

In practice

• Reuse prior task context for efficiency.
• Switch orchestrator models mid-task.

Topics

• Perplexity Computer
• AI Agents
• Contextual Memory
• Model Orchestration
• Website Publishing
• Private Company Data

Best for: CTO, VP of Engineering/Data, AI Architect, AI Engineer, Data Scientist, Director of AI/ML

Related on AIssential

• 😸 Perplexity Computer = CloudClaw? — Perplexity Computer orchestrates 19 specialized AI models to automate complex, multi-modal tasks in the cloud.
• Self-improving Memory for Agents - Perplexity — Brain enables AI agents to self-improve by learning from their past actions, successes, and failures, enhancing future task performance.
• How AI Agents Reshape Knowledge Work - Perplexity — AI agents like Perplexity Computer autonomously execute complex workflows, significantly reducing human effort and cost while expanding the scope and cognitive complexity of tasks.
• 🗞️ Claude Opus 4.7 launched as ‘less powerful’ version of Mythos — AI advancements are expanding into 3D world generation, desktop automation, and enhanced language models, while also raising concerns about cognitive impact and platform integrity.
• Perplexity Just Built an AI That Does Everything — Perplexity Computer offers a fully hosted, easy-to-use AI agent with extensive integrations, simplifying complex automation.
• Perplexity Computer bundles rival AI models into one agentic workflow system for $200 a month — Perplexity Computer unifies diverse specialized AI models into a single agentic workflow system.

See Counsel's argued verdicts on the open AI decisions leaders are weighing →

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by perplexity.ai via Google News.