Google Expands Gemini for Home Conversational Memory to 15 Minutes, Brings Gemini Live to Older Speakers
Google is rolling out a July 2026 update to Gemini for Home that extends conversational memory to 15 minutes across the entire experience and brings Gemini Live to the first-generation Google Home Mini and Nest Hub. The update ships alongside Google Home app version 4.22 and new Nest Cam firmware focused on connectivity and reliability fixes.
Google has expanded the conversational memory window for Gemini for Home to 15 minutes, according to a July 23, 2026 announcement from 9to5Google. The update means users no longer need to repeat context when issuing follow-up voice commands within that timeframe.
Google describes the change as applying "across the entire Gemini for Home experience." A practical example: a user can say "Hey Google, turn on the kitchen lights," then minutes later follow up with "Hey Google, dim it to 50%" without re-specifying which lights they mean.
The company did not disclose the previous memory window duration, making it unclear how large an increase this represents.
Gemini Live Reaches Older Hardware
Alongside the memory expansion, Google is bringing Gemini Live — its back-and-forth conversational voice mode — to the first-generation Google Home Mini and Nest Hub. Users activate it by saying "Hey Google, let's chat." The feature was previously available on Nest Mini, Nest Audio, Nest Hub Max, and Google Home Speaker, extending Gemini Live's reach to some of Google's oldest still-supported smart speaker and display hardware.
Google Home App 4.22 and Nest Cam Firmware
The update coincides with Google Home app version 4.22 for Android and iOS, which adds several changes:
- Automation alerts: Warnings when saving automations that have conditions but no starters
- Smart blind controls: Updated labels and controls for setting precise shade positions
- Smart lock disclaimers: Reminders that passcodes for removed household members must be manually deleted
- iOS video fixes: Lower playback error rates and improved near-real-time stream availability
- Playback resume fix: Corrected an issue where in-progress video feeds failed to resume after unlocking a phone
- Deep link fixes: Corrected navigation so setup and settings links open the right screen
- Battery alert fix: Fixed a bug reporting fully charged camera batteries as having zero hours remaining
Google is also rolling out a Nest Camera July 2026 software update for the Nest Cam Indoor (wired, 3rd gen), Nest Cam Outdoor (wired, 2nd gen), and Nest Doorbell (wired, 3rd gen). That firmware targets more reliable connectivity, a faster and more consistent setup flow, corrected Activity Zone alerts (fixing false person/familiar-face triggers inside ignored zones), and fixes for app freezes and incorrect camera status display.
Google has not specified a rollout timeline or percentage of eligible devices currently receiving either update.
What this means
This is an incremental quality-of-life update rather than a new model or architecture. The 15-minute memory window addresses a persistent complaint about voice assistants: they forget context almost immediately, forcing users to restate commands. Extending Gemini Live to first-generation Home Mini and Nest Hub hardware also suggests Google's Gemini-based assistant stack is now efficient enough to run acceptably on older, lower-powered edge devices — a meaningful signal for how Google plans to deploy conversational AI across its full installed base rather than just flagship hardware. The bundled Nest Cam fixes (connectivity drops, Activity Zone false positives, app freezes) address longstanding reliability complaints rather than adding new capabilities. None of these changes involve new benchmarks, pricing, or model architecture disclosures — this is a maintenance and UX release, not a capability leap.
Related Articles
AWS Launches Agentic Retrieval for Bedrock Knowledge Bases, Priced at $4 per 1,000 Calls
Amazon Bedrock Managed Knowledge Bases now offers agentic retrieval through a new AgenticRetrieveStream API that decomposes multi-part questions into sub-queries and iterates until it judges evidence sufficient. The managed model costs $4 per 1,000 agentic retrieval calls plus $1 per 1,000 underlying Retrieve API calls.
OpenAI Relaunches ChatGPT Health with Apple Health Integration for All US Users
OpenAI is relaunching its ChatGPT Health feature, expanding access to all US-based logged-in users 18 and older after a lackluster initial pilot that began in January 2026. The feature connects to Apple Health and medical records to provide personalized health context in ChatGPT conversations.
Anthropic Adds Opus and Sonnet Models to Claude Voice Mode, Ending Haiku-Only Limit
Anthropic has upgraded Claude's voice mode to run on Opus and Sonnet models for the first time, moving beyond the instant-response Haiku model it previously relied on exclusively. The update adds mid-conversation model switching, connector support, and coverage for 11 languages.
Google Expands Gemini Spark Agentic Assistant to All AI Pro and Ultra Subscribers
Google is expanding access to Gemini Spark, its agentic AI assistant built on Gemini 3.5, to all Google AI Pro subscribers in the US and Google AI Ultra subscribers globally. The rollout excludes free-tier users and, for Ultra, customers in the EEA, Switzerland, the UK, and Nigeria.
Comments
Loading...