Changelog

Changelog

What each update brings — only the changes you'll actually notice.

1. 1.0.12026-10-27
   Fixed

   Image generation and transcription now work on computers whose user name has non-English characters.

2. v00122026-10-23
   NewModel library

   MiMo V2.6 Distill 9B joins the model library.

   • A 9B model built for codingStrong at coding and multi-step tasks, and it can read images. Download it from the model page and it is ready to use, running entirely on your computer.
3. 1.0.02026-10-23
   New

   1.0 turns six tools into one workspace, with a Library and live transcription.

   • Library and hand-off between toolsEverything you make lands in one Library: search it, pin it, tag it, or send it to another tool to keep working.
   • Live transcription and speaker separationLive transcription (Beta) shows captions while you record; transcripts can tell speakers apart, and exports and insight reports carry the names.
   • API keys and AI agent connectionsDevices on your network use this computer's models with a key, and AI agents can call text recognition and speech transcription.

   Read the full overview →
4. 0.63.72026-09-17
   Fixed

   Models now start on computers that use a proxy, and fixed-name connections are more reliable.

5. 0.63.62026-09-16
   Improved

   Local network access can use a fixed name, plus fixes for known issues.

   • Connect by a fixed nameTurn it on in Settings and other devices connect by a fixed name, so a changed network address never means a changed URL. The name prefix is yours to choose.
6. 0.63.52026-09-05
   Fixed

   Fixes API endpoint issues and a settings page display issue.

7. 0.63.32026-09-01
   Improved

   Hardware diagnostics show your graphics acceleration status and remind you when the driver needs updating.

8. 0.63.22026-08-31
   Fixed

   Local network access is easier to use, and conversation length now fits your computer automatically.

   • Local network accessTurn it on and the addresses you can connect to are listed for you. These connections need no password, so use it only on networks you trust.
9. 0.63.12026-08-26
   New

   Chat roles are yours to create and change.

   • Change a role any timeChange its name, description and wording without starting over. You can also copy a built-in role and adjust it.
   • See where your wording goesThe management view shows where your wording sits in the conversation.
10. 0.63.02026-08-25
    New

    Conversations show how much room is left, and the settings explain themselves.

    • See the room you have leftA ring next to the input box shows how much this conversation has used. Start a new one when it gets close to full.
    • Model details are accurateThe length shown in the model list is the model's own limit, so you can pick by it directly.
    • Advanced settings warn you firstEvery field states its default and range, and tells you the risk before you change it.
11. 0.62.02026-08-21
    New

    Qwen 3.8 27B joins the model library.

    • Qwen 3.8 27B is availableDownload it from the model page and it is ready to use. It reads images and handles very long documents, and like the others it runs entirely on your computer.
12. 0.61.12026-08-18
    New

    Speech transcription supports more audio and video formats.

13. 0.61.02026-08-17
    Fixed

    Improves speech transcription output.

14. 0.60.22026-08-12
    Fixed

    The model list keeps up with new models on its own, and downloads say what they will fetch.

15. 0.60.12026-08-12
    New

    Muse Glimmer 30B joins the model library.

    • Muse Glimmer 30B is availableDownload it from the model page and it is ready to use. It reads images and reasons, and like the others it runs entirely on your computer.
    • Engine updates in one clickWhen an update is needed the app tells you first, and one click updates it and starts the model.
16. 0.59.02026-08-10
    Improved

    Transcripts follow your interface language, and installation is steadier.

17. 0.58.02026-08-05
    Improved

    Chat answers are more dependable, and image generation is steadier.

    • More dependable answersAnswers use the interface language you picked, and say so plainly when something is uncertain instead of inventing figures.
18. 0.57.02026-07-30
    New

    Attach a document to a chat and ask about it.

    • Ask about a documentAttach a PDF, Word or text file and ask away, long ones included. Answers come with the passages they rest on.
    • Works for insights tooResults from text recognition and the speech workshop can be questioned the same way.
19. 0.56.02026-07-28
    New

    Adds image upscaling, and image generation is faster.

    • Image upscalingUpload a picture to enlarge it and bring detail back. No regenerating, no prompt needed.
20. 0.55.02026-07-27
    Improved

    The chat and insight screens have been reorganised and read more easily.

Only updates that affect you are listed, so version numbers may skip.