Mmory launches private on-device memory app

// 2h agoPRODUCT LAUNCH

Mmory launches private on-device memory app

Mmory is an iPhone app for recording or typing thoughts, then retrieving them with natural-language questions. Transcription, search, answers, and reminders run locally on the device.

// ANALYSIS

Mmory’s strongest pitch is privacy without sacrificing the convenience of AI recall, though its iPhone-only, English-only scope limits reach.

–On-device processing keeps sensitive memories and voice recordings off Mmory’s servers
–Natural-language retrieval is more useful than replaying or manually searching voice memos
–Reminders turn passive memory capture into an actionable workflow
–The 50-save free tier and $3.99 monthly Pro plan create a clear upgrade path
–Siri integration is still planned, leaving a major friction point for hands-free capture

// TAGS

mmoryspeechsttagent-memorylocal-firstsearch

DISCOVERED

2h ago

2026-11-06

PUBLISHED

7h ago

2026-11-06

RELEVANCE

6/ 10

AUTHOR

Dhruv Dadoo

// KEEP READING

More AI developer news from the feed

EXPLORE FULL FEED

RESEARCH35m ago

Coca-Cola Compresses Tool Outputs Into Latents

Tsinghua researchers’ Coca-Cola replaces verbose tool observations with continuous latent tokens injected directly into an agent’s embedding space. Its action-preserving training targets next decisions rather than text reconstruction, reducing token consumption by 70.01%-89.24% and runtime by 24.45%-55.97% across web-search QA benchmarks.

BENCHMARK1h ago

Liber-0 Hits 36% Zero-Shot Robot Success

LiberAI’s World Action Model achieved 36% success across 600 trials on 60 GPT-6-generated manipulation tasks without task-specific robot demonstrations. It reached 63.5% on pick-and-place, 34% on short manipulation, and 10.5% on challenging multi-step tasks. ([LiberAI](https://www.liberai.com/blog/liber0/))

RESEARCH1h ago

Route to Remember Makes Weight Memory Practical

Amazon researchers show that on-policy self-distillation can make per-user weight memory outperform text retrieval for factual recall, reaching 59.9% accuracy versus 47.8% for retrieval on HaluMem. A verify-then-regenerate router combining both approaches raises accuracy to 66.1%.