v0.6.11

v0.6.11

The goal of this update was to make the primary menu more accessible via only thumb use on the bottom bar.

Features:

• Added proper back button handling for the Settings Drawer, gear Menu and return to primary menu.
• Added a clipboard paste button to the import Character modal.

Changes:

• Moved Regenerate and Continue to the Swipe bar.
• Moved Edit Character and Chat History to gear Menu.
• Removed return to primary menu button, as its now done with the system back button.
• Improved TTS logic to not cause cutoffs for long texts.
• Changed Swipe Bar buttons to be a lot bigger and easier to click.

Fixes:

• You should be able to interact with ChatItem buttons while the keyboard is up.

v0.6.10

v0.6.10

Features:

• Pushed llama.rn version. Google Gemma models now supported!
• Added model selector to generic Text Completions backend.
• Added a 'Recently Opened' menu on the Welcome screen.
• Added a button to return to the Welcome screen from a chat.
• Added an option to automatically trigger TTS when a generation finishes.

Changes:

• Settings Menu is now a Drawer component.
• Updated all screen transitions and added animations to most components.
• Newly generated text should be properly trimmed.
• Updated Local inferencing logs to be more readable.

Fixes:

• Fixed Horde responses failing due using an async payload builder.
• Fixed character portraits not loading properly.
• Fixed local inferencing breaking due to double saving.

v0.6.9

v0.6.9

Features

• Added downloading from character cards from pygmalion.chat
• Updated llama.rn to latest
• Added a version tag to the settings menu.

Fixes

• Updated how character portraits are handled in chat to properly use fallback image.

Changes

• Refactored many components to be more maintainable, this is mostly for development so it shouldn't change anything.

v0.6.8

v0.6.8

Features:

• Added a Logger that will log app state changes.
• Added autosaving to Instruct and Character Card editors.
• Added OpenAI backend!

Changes:

• Migrated Message state to Zustand, this should hopefully improve performance.
• Updated default avatar handling.
• Translated many files to Typescript for better error handling and safety.

v0.6.7

v0.6.7

• Updated messages to use FlatList to reduce re-rendering
• Added small delay when loading characters to properly load chat history
• Made message management improvements on the backend, nothing should have changed for the end user here

v0.6.6

v0.6.6

• Fixed issues described in #5.
• Fixed system sequence not being used.
• Prettier-fied code and now implemented proper linting.
• Bumped to Expo SDK 50

Going forward, releases will now be on every minor version, instead of updating the major release.

v0.6

v0.6

New feature: Added llama.rn backend to run models natively.

• Simply download a model to your phone and Import it. (Note that it creates a duplicate of the model file within the app, so it may be wise to remove the downloaded model to save on space.)
• Load the model to allocate it into memory and start chatting.
• Prompt processing is very slow, expect it to take a few minutes on a new chat.
• llama.rn does not have GPU support as of yet, so prompt processing will be very slow.

v0.6.1

• Initial implementation of Open Router and Chat Completions, currently does not respect Instruct formatting.
• Added System TTS

v0.6.2

• Implemented alternate greetings.
• Fixed issues with macro replacement.

v0.6.2a

• Fixed macro replacement still not working.

v0.6.3

• Optimized context building.
• Added elipses when waiting for response.
• Improved startup routine.

v0.6.3a

• Fixed issues with local inferencing.

v0.6.4

• Added dynatemp for KAI endpoint.
• Allow newlines in Instruct presets.

v0.6.4a

• Fixed a crash when entering CharInfo menu.

v0.6.5

• Bumped llama.rn version to 0.3.0-rc.14
• Added Seed value to local inferencing.

v0.5

v0.5

This release is mostly housekeeping and making the codebase less messy.

• Added Text Completions support (not fully tested, might be broken).
• Unified Presets.
• Changed structuring of global values.

v0.5.1

• Fixed an issue with textgen max tokens using incorrect field.
• Added min_p to textgen and Horde.

v0.5.2

• Fixed issues with KAI responses being improperly formatted.

v0.5.3

• Fixed responses not being most recent messages in chat.

v0.5.4

• Fixed the Instruct menu being completely broken.

v0.5.5

• Fixed Instruct labels being missing.
• Fixed modifying User personas crashing the app.
• Fixed placeholders for editing chat items not being updated upon generation.

v0.5.6

• Fixed new chat creation causing a crash.
• Fixed app initialization not creating default preset and instructs.

Adventure Test Release

For testing adventure mode.

#v0.4.1-a Hotfix

• Fixed crash when launching with adventure mode enabled.

v0.4

v0.4

• General improvements and fixes, mostly clean up from v0.3
• Added chat number, generation time and replaced the chat bubble edit button with a hold to edit feature.

v0.4.1

• Fixed improper instruct formatting.