Repository Issue Activity (beta feature)

#2056 Gemma 4 E4B SIGSEGV in CPU vision path on Android (second image turn), v0.10.0 & v0.10.2

#1931 litert-lm-api (0.10.1) has no wheels for android_30_arm64_v8a

#2031 Vulnerability in LiteRT-LM project

#2028 SIGSEGV in liblitertlm_jni.so on second createConversation() when Engine runs in an android:process=":ai" Foreground Service (iQOO I2407, CPU, Gemma 4 E2B)

#1992 Add SamplerConfig support to the Python API (parity with Kotlin)

#1377 Unable to run LiteRT-LM on Qualcomm device

#1972 About the kotlin version

#547 Converting Gemma3n to litertlm

#2020 [BUG] FastVLM 0.5 B Native Crash (SIGABRT) during NPU Initialization on Snapdragon 8 Elite (SM8850)

#2018 Unable to benchmark Gemma-4-E2B-it using `litert_lm_advanced_main`

#2001 GPU backend support in litert-lm-api on linux_aarch64

#2003 How to build litertlm-android.aar manually?

#1847 litert-lm CLI binary fails with "illegal hardware instruction" on Raspberry Pi 4

#1998 Error fixes

#1973 Publish releses on maven.google.com

#1999 GPU backend on macOS from Python produces a lot of log (Python API)

#1466 Running into Dispatch API issue while executing Sample App via Android App

#2002 [Showcase] Watcher uses LiteRT-LM in an Android AI video app

#1959 [Community] LiteRTLM-Swift — Swift Package wrapping the C API for iOS

#2000 [Windows] JVM SDK fails for models > 2 GB — stat() in nativeCreateEngine / nativeCreateBenchmark hits MSVCRT 32-bit overflow

#1864 Gemma-4-E4B-it and the E2B-it variant fail to Initialize on Samsung Exynos 2600 — LiteRT Engine Error

#1993 `litert_lm_cli serve` lets one slow request block all others due to single-threaded `HTTPServer`

#1990 libLiteRtTopKMetalSampler missing from prebuilts / no Bazel target on iOS (Metal backend falls back to CPU sampling)

#1979 Inference cannot be performed using NPU on 8550 SoC

#1050 [Feature Request] Enable building GPU accelerator plugins from source for cross-platform distribution