Back to News

Perplexity launches Hybrid Compute for Mac app, mixing cloud and local LLMs

#perplexity#hybrid compute#mac app#local llm

Perplexity has launched Hybrid Compute for its Mac app, available to all users, which splits workloads between frontier cloud models such as Opus 5 and local LLMs. The feature aims to leverage local processing, potentially reducing costs.

Coverage timeline

  1. Techmeme
    Perplexity launches Hybrid Compute, which splits workloads between frontier cloud models like Opus 5 and local LLMs, for all users of its Mac app (Igor Bonifacic/Engadget)

    Igor Bonifacic / Engadget : Perplexity launches Hybrid Compute, which splits workloads between frontier cloud models like Opus 5 and local LLMs, for all users of its Mac app — And local processing could keep costs down too. — Back in February, debuted Perplexity Computer. Like Claude Cowork …

Related stories

  • Z.ai confirms Ox Alpha is GLM-5.3 Flash, releases open weights

    Z.ai confirmed that the anonymous model 'Ox Alpha', which went viral on OpenRouter, is a new iteration of its GLM series, specifically GLM-5.3 Flash, and released its weights on Hugging Face under a new license. The model, a 320B-parameter natively multimodal model running on domestic chips, topped OpenRouter's leaderboard and ended DeepSeek's 56-day streak on OpenCode. The license requires companies with over $10B in revenue to pass a security review.

    9 reports · 4 sources801 points · 281 comments

  • Study Characterizes Massive Activations in Hybrid Linear Attention LLMs

    Researchers from StartLux, Tsinghua University, and other institutions published a study using massive activations (MAs) as probes to systematically characterize cross-layer traces of full attention layers in hybrid linear attention LLMs, identifying two core patterns: pre-attention spikes (PAS) and inter-spike plateaus (ISP). The study covers five linear attention architectures, six hybrid configurations, five data domains, and 12 public HLA LLM checkpoints, with model sizes ranging from 1.2B to 397B parameters.

    1 reports · 1 sources

  • LLM 0.32 released with reasoning traces, OpenAI Responses, content-addressable logs

    Simon Willison released LLM 0.32, a major update to his command-line tool for interacting with large language models. The release adds support for visible reasoning traces, server-side provider tools, and a redesigned content-addressable SQLite log system that de-duplicates stored messages. It also introduces new default models and features enabled by the OpenAI Responses API, alongside an updated llm-anthropic plugin.

    4 reports · 1 sources

  • Meta launches Mac app for Meta AI with dictation, screen context and business integrations

    Meta has launched a new Mac app for Meta AI, featuring system-wide dictation and the ability to answer questions based on current screen content using its Muse Spark model. The update, aimed at business owners, lets merchants connect Instagram and Facebook accounts, Meta ad campaigns, and Google Workspace to query campaign performance and audience engagement insights.

    2 reports · 2 sources

News updates every hour. Subscribe in your RSS reader to get new stories automatically.

Subscribe