v0.10.1

🔥 Gemma 4 support

Deploy Gemma 4 across a broad range of hardware with stellar performance (blog).

👉 Try on Linux, macOS, Windows (WSL) or Raspberry Pi with the
LiteRT-LM CLI:

litert-lm run  \
   --from-huggingface-repo=litert-community/gemma-4-E2B-it-litert-lm \
   gemma-4-E2B-it.litertlm \
   --prompt="What is the capital of France?"

Release Notes

• CLI Enhancements & Migration: Migrated the CLI from fire to click, adding features like --verbose, --version, improved help formatting, and enhanced terminal output styling (#1784, #1733, #1791, #1792).
• Hugging Face Integration: Added support for importing models directly from Hugging Face and implemented auto-conversion for missing models during "run" commands (#1797, #1735).
• Core Performance & Features: Introduced a LiteRT-based KV cache implementation, speculative decoding support, and improved context merging for conversation history (#1601, #1793, #1742).
• Platform & Build Improvements: Refactored CMake for better Android/cross-compilation support, updated the Windows build with a CPU sampler workaround, and transitioned nightly releases to Ubuntu-22.04 (#1741, #1734, #1772).
• API & Documentation: Expanded the Kotlin API for response channel configuration and launched new Python API resources, including a "Getting Started" guide and a Colab notebook (#1724, #1737, #1757).

v0.9.0

Android & iOS Update

• Performance Optimizations: Significant improvements to app initialization speed and memory management.

• Bug Fixes: General stability enhancements for a smoother user experience.

v0.9.0-rc

Android / iOS release
With many bug fixes and performance improvements.

v0.9.0-beta

Beta release for v0.9.0.

v0.9.0-alpha03

Update to alpha03 for 0.9.0 release

v0.9.0-alpha02

Update to alpha02 for 0.9.0 release

v0.9.0-alpha

Support multi-session and session-cloning for compiled model executor.

LiteRT-LM-PiperOrigin-RevId: 862916621

v0.8.1

Bugfix:

• Fix the Windows compilation error when using GPU

V0.8.0

• Desktop GPU support.
• Simple CLI for Desktop: Link to Quick Start section
• Multi-Modality support: Vision and Audio input are supported when models support it. See more details here
• Kotlin API for Android and JVM (Linux, MacOS, Windows): Link to LiteRT-LM Kotlin API
• Conversation API: Link to Conversation API
• Function calling support: Link to Tool Use

v0.7.0

• Added Qualcomm & MediaTek NPU support for Gemma3 1B model!
• Fixed Apple M4 build issue.
• Other minor refactorization / cleanups.