Skip to content

Releases: orailnoor/cross-platform-llm-client

Release list

PrivateLM App v1.0.5

PrivateLM App v1.0.5 Latest
Latest

Choose a tag to compare

View all tags
@techjarves techjarves released this 21 Jul 01:33
1.0.5
df3b8c9

PrivateLM 1.0.5

This release addresses the following reported issues:

  • LiteRT loading now retries on CPU when GPU initialization fails (#11).
  • Cloudflare and ngrok tunnel binaries and integration have been removed (#13, #21).
  • Fixed the GetX/Obx streaming-chat runtime error (#17).
  • GGUF models can now be served through the local OpenAI-compatible API server (#19).
  • Model deletion now requires confirmation (#22).
  • Downloads are persisted and reconciled after leaving or restarting the app (#24).
  • Existing cloud-provider API keys can now be replaced (#25).
  • FP16 image-model loading support has been restored (#28).
  • Landscape orientations are supported (#30).
  • Text-only Gemma LiteRT models are no longer incorrectly classified as vision models (#33).
  • Multiple custom API URL profiles can be saved and selected (#34).
  • App version display now comes from package metadata.
  • GGUF native load errors include more useful diagnostics (#32; diagnosis improved, device-specific compatibility is not claimed fixed).

Android downloads

  • app-arm64-v8a-release.apk: most modern Android phones and tablets
  • app-armeabi-v7a-release.apk: older 32-bit ARM devices
  • app-x86_64-release.apk: x86_64 Android devices and emulators

Important upgrade/signing notice

The prior GitHub APK was signed with an unavailable debug certificate. Android therefore cannot install this release over that APK. Existing users must uninstall the previous build once before installing 1.0.5. Future in-place upgrades require the project owner to preserve and use one stable release keystore. Issue #27 remains open until that release-signing requirement is resolved.

SHA-256

a587dfba36d8832e07174afbb9ee699c91d82dce5d6b7b32b9aa8fd5b2f801c5  app-arm64-v8a-release.apk
483362313a4500c2c8fba65e2e30c1fbe3ff53b3b3430e2d9697dfcb81b4a221  app-armeabi-v7a-release.apk
dab04b5f18960ba7abde35dbb93afa4029f66d87fe8157a56a17b1bd49e304dd  app-x86_64-release.apk

Validation: targeted Flutter tests passed, the changed Dart files compiled without errors, and all three Android APK variants built successfully.

Assets 5
Loading

PrivateLM App v1.0.4

PrivateLM App v1.0.4

Choose a tag to compare

View all tags
@orailnoor orailnoor released this 27 Apr 09:42
1.0.3
1ab16e5

New Features

  • Document AI — Upload and analyze PDF and DOCX files. Text is automatically extracted and added to the conversation context.
  • Image Generation Progress — Added a real-time progress bar with percentage, step counter, and estimated remaining time during image generation.
  • Device Compatibility Advisor — Settings now detects your device chipset (Snapdragon, Tensor, Apple Silicon, etc.) and recommends the most suitable model quantization format.
  • Resume Session Dialog — On startup, the app now asks whether you want to reload previously used models instead of restoring them automatically.
  • Local Mode by Default — Fresh installs now default to on-device inference instead of cloud mode.

Bug Fixes

  • Fixed chat context leaking between conversations when switching chats.
  • Fixed incorrect “restart required” warnings when switching from LiteRT models to GGUF models.
  • Fixed empty responses on Pixel 6, 7, and 8 devices by running Gemma models in single-threaded mode on Google Tensor chips, along with automatic garbage-token filtering and clearer error handling.
  • Fixed temporary vision attachment files not being cleaned up after removal.
  • Fixed cross-chat message contamination when deleting or switching chats during active AI streaming.
  • Fixed multiple memory leaks across Server, Task, and Model views involving TextEditingController and StreamSubscription instances.

Improvements

  • Improved the New Chat button with a more intuitive edit_note icon and tooltip.
  • Replaced raw native image model crash dumps with actionable recommendations such as trying CyberRealistic, Realistic Vision, or AbsoluteReality.
  • Unsupported file formats are now blocked with a clear toast message instead of failing silently.
  • Added active image model indicators in the app bar.
  • The last used image model is now remembered across app restarts.

Removed

  • SDXS 512 — Removed from the image model catalog due to widespread compatibility issues across devices.

Full Changelog

https://github.com/orailnoor/cross-platform-llm-client/commits/v1.0.0

Loading