Anuma
What you can do
Chat with AICreate with AIBuild with AIAI Council
AI Model Access
Multiple AI ModelsOpen-Source LLMsClosed-Source LLMs
Nearby
Unified MemoryPrivacyPortable Context
Work & Career
ProfessionalsDevelopersEntrepreneursJob SeekersFreelancersSmall Business OwnersParents
Students & Learning
StudentsResearchersLanguage LearnersTeachers & Educators
Creative & Content
WritersDesignersContent CreatorsMusicians
Finance & Legal
Personal FinanceLegal & ContractsInvestment ClubsReal Estate
Help CenterWhat's NewBlogFAQAI Prompt LibraryHow Memory WorksOpen-Source LLMsClosed-Source LLMsPrivate by Default
Pricing
Get the appTry Anuma
Get the app
Anuma
What you can do
Chat with AICreate with AIBuild with AIAI Council
AI Model Access
Multiple AI ModelsOpen-Source LLMsClosed-Source LLMs
Nearby
Unified MemoryPrivacyPortable Context
Work & Career
ProfessionalsDevelopersEntrepreneursJob SeekersFreelancersSmall Business OwnersParents
Students & Learning
StudentsResearchersLanguage LearnersTeachers & Educators
Creative & Content
WritersDesignersContent CreatorsMusicians
Finance & Legal
Personal FinanceLegal & ContractsInvestment ClubsReal Estate
Help CenterWhat's NewBlogFAQAI Prompt LibraryHow Memory WorksOpen-Source LLMsClosed-Source LLMsPrivate by Default
Pricing
Get the appTry Anuma
Get the app
Back to What's new

Save any line as a memory

May 18, 2026·2 min readMemoryNewStreamingMobile
Save any line as a memory

Highlight, save, ask

When something in a chat reply is worth keeping, you can now highlight it and a small toolbar appears under the selection. Save to memory rewrites the selection into a clean entry — you can edit it inline before adding it to your Memory Vault. Ask Anuma captures the selection as a quote chip above the composer, so your next message carries the context as a markdown blockquote. The toolbar works on both assistant and user messages, and morphs in place between idle, generating, and preview phases.

A smoother stream

The streaming markdown pipeline got rebuilt this week. Previously-streamed paragraphs no longer flash on every commit boundary, layout doesn't shift by 16–24px when a new block appends, code blocks don't render full-height before snapping shorter, and mid-stream math survives the final render. New tokens fade in by word; everything before stays stable.

Try Anuma before signing in

Visitors landing on Anuma without an account now get a real chat surface — full sidebar, suggestion chips, Council Mode, and three free messages before the sign-in prompt. Locked surfaces (non-Auto models, file upload, image and video generation, deep research) open a clear sign-in dialog with a feature-specific headline instead of a blank wall.

Storage you can see

A new Storage section in Settings shows real on-device usage with a per-category breakdown — images, videos, audio, documents, slide decks, and apps. The header reads actual browser-reported usage and quota, with a pill progress bar that turns yellow at 75% and red at 90%.

On mobile

The mobile rebuild kept pace this week. The Library got a 2-column grid, full-screen image and video viewers, inline WebView for PDFs and HTML, and an always-on auto-save so generated content lands in the library automatically. Recents picked up a hybrid keyword + semantic search with a 300ms debounce. Long-press any artifact in chat or library to open a unified Share / Save / Edit / Copy / Open sheet. The Agents section got a full redesign — large title, recently-used orb row, filter pills, and richer agent cards. Encrypted backup is now on by default for new users. A new app icon shipped. And a big backup OOM fix flattened a 168 MB heap climb to a steady ~97 MB across nearly 1,000 ops.

← Back to What's new

What you can do

  • Chat
  • Create
  • Build

Core features

  • Unified Memory
  • Multiple AI Models
  • Council Mode
  • Private by Default

Solutions

  • Work & Career
  • Student & learning
  • Creative & Content
  • Finance & Legal

Company

  • About
  • Careers
  • Branding
  • Contact us
  • Affiliate

Resources

  • Help Center
  • Blog
  • FAQ
  • Prompt Library
  • How memory works
  • Open-Source LLM
  • Closed-Source LLM
iOSAndroid
Anuma
Powered by
© 2026 Anuma, Inc. All rights reserved.|Privacy|Terms|Cookie Policy||