chore(deps): bump flutter_gemma from 0.3.1 to 0.15.3

[dependabot]

Bumps flutter_gemma from 0.3.1 to 0.15.3.

Release notes

Sourced from flutter_gemma's releases.

Release v0.15.3

• example: add TranslateGemma 4B translation demo via task-first home navigation (#177).

CPU-only by design — community .litertlm keeps EMBEDDING_LOOKUP in float32 for MediaPipe .task compatibility, and the LiteRT GPU partitioner can't cluster that layout. Tracked upstream at google-ai-edge/LiteRT-LM#1748.

Release v0.15.2

Unified embedding on LiteRT C API + Dart FFI on all native platforms (#264).

What changed

• Single Dart-side LitertEmbeddingModel runs on Android, iOS, macOS, Linux, Windows via dart:ffi against the LiteRT C API. The LiteRT runtime is already shipped in the native bundle for inference accelerators — no new native dep.
• Three legacy implementations deleted: Android localagents-rag JVM lib, iOS TensorFlowLiteC pod + Swift tokenizer hierarchy, Desktop libtensorflowlite_c.{dylib,so,dll} tarball.
• TaskType.prefix is now the single source of truth for retrievalQuery / retrievalDocument prefix strings — fixes the cross-platform drift that caused #264 by construction.
• Windows LiteRtLayout MSVC vs GCC/Clang bit-field ABI handled via dual-layout FFI structs (LiteRtLayoutPosix / LiteRtLayoutMsvc + View helper).

Verification

Cross-platform RAG drift verified <0.003 on six platforms × two embedding models (Gecko 110M, EmbeddingGemma 300M):

	macOS	iOS Sim	iPhone	Pixel 8	Linux T4	Windows MSVC
Gecko 64 climate↔warming	0.9161	0.9161	0.9161	0.9168	0.9163	0.9163
EmbeddingGemma 256 climate↔warming	0.8695	0.8695	0.8695	0.8689	0.8668	0.8668

Full 18-test litertlm_ffi_test.dart integration suite green on macOS, Linux T4, Windows NPU VM (Intel Lunar Lake), Pixel 8.

Breaking changes

• EmbeddingInstallationBuilder.tokenizerFromNetwork/Asset/Bundled/File: removed the iOS-only iosPath / iosToken parameters. They were no-ops since 0.15.2 (same tokenizer on every platform now). Callers that omit them are unaffected; callers that pass them need to remove the named arg.

See CHANGELOG.md and the migration story in the PR description for the full diff.

Release v0.15.1

What's new

• 🪟 Windows Intel NPU end-to-end — PreferredBackend.npu on LunarLake/PantherLake. Native bundle ships Intel dispatch + OpenVino + TBB.
• 🖼️ Multi-image input (PR #262, thanks @​frdteknikelektro) — Message.withImages([...]).
• 🤖 Android GPU sampler dlopen fix (#270, thanks @​prithidevghosh) — +15–30% decode speedup on Pixel-class devices.
• 💾 Desktop storage (#179, co-author @​ProjectEdge-Jim) — Application Support / LOCALAPPDATA instead of Documents.
• ⚙️ NPU greedy-only sampler skip on PreferredBackend.npu.
• 🌐 Web build fix — enableSpeculativeDecoding parameter missing in FFI stub.
• 📦 native-v0.11.0-b — new Windows tarball with Intel NPU; other 6 platforms unchanged.

Test plan

• macOS Metal: Gemma 4 E2B full suite 17/17
• Pixel 8 Android 16: Gemma 4 E2B full suite 17/17
• Windows LunarLake VM: NPU end-to-end "Paris" via PreferredBackend.npu

See CHANGELOG.md for full details.

native-v0.11.0-b

Windows: bundles Intel NPU dispatch (LiteRtDispatch.dll + OpenVino runtime + TBB, 12 extra DLLs) for PreferredBackend.npu on LunarLake-class Intel chips. Built from LiteRT-LM commit 62f7a8e. Other 6 platforms unchanged from native-v0.11.0-a (032334d).

... (truncated)

Changelog

Sourced from flutter_gemma's changelog.

0.15.3

• example: add TranslateGemma 4B translation demo via task-first home navigation (#177).

0.15.2

• Unified embedding on LiteRT C API + Dart FFI on all native platforms (#264).

0.15.1

• Fix Android GPU sampler dlopen failure (#270, thanks @​prithidevghosh): patchelf --add-needed libLiteRtLm.so on libLiteRtTopK{OpenCl,WebGpu}Sampler.so.
• Desktop storage (#179, co-author @​ProjectEdge-Jim): use Application Support instead of Documents on Windows/macOS/Linux to avoid cloud-synced paths breaking FFI mmap.
• Multi-image, FFI session metrics, prefix replay (#262, thanks @​frdteknikelektro): Message.withImages([...]) for multi-image input, chat/Conversation session metrics via FFI, persistent prefix messages replayed on session rebuild after history truncation. Backward-compatible (Message.withImage(...) still works).
• Skip sampler params on NPU backend: temperature/topK/topP/seed are silently ignored when PreferredBackend.npu is selected — LiteRT-LM NPU executor only supports internal greedy sampling.
• Windows Intel NPU end-to-end: native-v0.11.0-b Windows tarball bundles Intel dispatch (LiteRtDispatch.dll), OpenVino runtime, TBB (~30 MB); FFI client passes dispatch_lib_dir + use_hw_masking_for_npu=false at engine_create so PreferredBackend.npu works on Intel LunarLake/PantherLake silicon without manual DLL placement.
• Fix web build broken by 0.15.0: LiteRtLmFfiClient stub on web was missing the enableSpeculativeDecoding parameter — dart2js failed compilation when web target was actually built.
• CI: standalone build-litertlm-native-windows.yml workflow for Windows-only rebuilds.

0.15.0

• LiteRT-LM 0.11.0: MTP-capable Gemma 4 + speculative decoding on macOS / iOS / Android / Windows.
• enableSpeculativeDecoding flag on getActiveModel() (null = model default; true/false to override).
• Restore Android NPU support for .litertlm (regression from 0.14.0): PreferredBackend.npu on Android .litertlm models routes through LiteRT-LM's Backend::NPU again — same as 0.13.x's Kotlin path before the FFI migration silently dropped it. Requires a Qualcomm QNN / Google Tensor / MediaTek dispatch lib on the device; without one, engine_create fails with a dispatch error. MediaPipe .task models still don't support NPU (MediaPipe SDK limitation, unchanged).
• PreferredBackend.npu on desktop (#261): macOS / Linux / Windows backend arm wired. Same dispatch-lib requirement as Android.
• Linux known limitation: post-MTP HF Gemma 4 revisions blocked upstream — libLiteRtWebGpuAccelerator.so segfaults during graph compile (filed google-ai-edge/LiteRT-LM#2225) and multi-signature vision encoder rejected by native (engine_create error). Workaround: pin pre-MTP HF revision 7fa1d78473894f7e736a21d920c3aa80f950c0db for Linux users. Other platforms unaffected.
• hook native cache: marker-file based invalidation (.flutter_gemma_native_version) — bumps clean stale companion libs automatically.

0.14.5

• Fix desktop embedding on pub.dev installs (#250 follow-up): tensorflowlite_c.{dll,so,dylib} now bundled via Native Assets — regression from 0.14.0 setup-script removal.

0.14.4

• Fix macOS dylib loading on pub.dev installs (#255).
• Fix fromAsset install on desktop (#250 mode 2).
• Fix Android x86_64 emulator crash on embedding init (#250 mode 3).
• Reduce native lib size: iOS -63%, macOS -43%, Android -28%, Linux -16-18%.

0.14.3

• Fix App Store ITMS-90208 rejection on iOS (#245): downgraded patched libGemmaModelConstraintProvider.dylib minos 26.2 → 14.0 to match the other companion dylibs.
• Fix Android Google Play 16KB page-size requirement (#253): rebuilt libLiteRtLm.so with -Wl,-z,max-page-size=16384.
• Fix Web RAG build pipeline (#251): cross-platform npm run build, bundles cache_api.js + LiteRT WASM into dist/, ships package.json + vite.config.js in pub tarball.

0.14.2

• Fix macOS install_name_tool failure (#247): dart run / build_runner / flutter test on a pure-Dart library aborted with larger updated load commands do not fit because upstream Apple companion dylibs lack -headerpad_max_install_names. Skip them from Native Assets on macOS and bundle via Podfile post_install instead.
• Fix Gemma 4 escape-token leakage in chat history (#248): <|"|> tokens from tool_calls.arguments were written to history raw, making the model echo them on later turns. Strip recursively before persist via new SdkResponseParser.cleanRawForHistory.
• Fix modelFromAsset install on desktop (#250): the large_file_handler channel call threw MissingPluginException on macOS / Windows / Linux because the package only ships Android + iOS plugins. Now catches it and falls back to the in-memory loadAsset → writeFile path. Embedding (localagents-rag) and .litertlm FFI paths additionally fail fast with a typed message on non-arm64 Android (x86_64 emulator / armeabi-v7a) instead of a generic JVM crash — see new "Platform & Architecture Support" section in README.
• macOS Podfile setup (BREAKING for macOS app projects): app projects must update macos/Podfile post_install to the new snippet (see README → macOS Setup). iOS / Linux / Windows / Android unaffected.
• Perf debug logs: [*/perf] lines break out cost of dylib load, engine_create, prefill, decode — visible via debugPrint in both debug and release builds.

0.14.1

• Web build fix (#244): 0.14.0 broke web compilation by statically importing core/ffi/litert_lm_client.dart (which imports dart:ffi) through the flutter_gemma_interface.dart → mobile/flutter_gemma_mobile.dart chain. JS/Wasm targets cannot compile dart:ffi. Added conditional imports in mobile/flutter_gemma_mobile.dart that swap litert_lm_client.dart and ffi_inference_model.dart for *_stub.dart shims on dart.library.js_interop. The web plugin (FlutterGemmaWeb) registers itself as FlutterGemmaPlugin.instance before any FFI code path runs, so the stubs' constructors (which throw UnsupportedError) are never actually invoked on web.
• App Store fix (#245, ITMS-90432): 0.14.0 dropped lib*.dylib symlinks alongside the bundled .framework/ accelerators in Runner.app/Frameworks/ so LiteRT-LM's gpu_registry could dlopen them by basename — App Store Connect rejected those builds with "Unexpected file found in Frameworks". 0.14.1 patches the upstream LiteRT-LM source (runtime/components/sampler_factory.cc and litert/runtime/accelerators/gpu_registry.cc) to load Apple platform accelerators via @executable_path/../Frameworks/<X>.framework/<X> (macOS) / @executable_path/Frameworks/<X>.framework/<X> (iOS) instead of libX.dylib. Native Assets bundles the framework bundles correctly out of the box; no host-side Podfile symlinks needed. Patch applied during local Bazel rebuild — see native/litert_lm/patch_c_api.sh section 10.
• Gemma 4 native function calling: New ModelType.gemma4 routes tool definitions to LiteRT-LM SDK via litert_lm_conversation_config_set_tools (OpenAI Chat Completions JSON). SDK applies chat_template.jinja through minja, renders native <|tool>declaration:...<tool|> tokens, and parses the model's <|tool_call>...<tool_call|> response back into structured tool_calls JSON. flutter_gemma reads the result via SdkResponseParser.extractToolCalls (handles parallel calls and the multimodal content[] path) and returns FunctionCallResponse to the app — no Dart-side prompt engineering needed.
• Escape token cleanup: Strips leftover <|"|> Gemma 4 escape tokens from string arguments (recursively, including nested maps/lists).
• example/lib/models/model.dart: Gemma 4 E2B / E4B entries switched to modelType: ModelType.gemma4.

... (truncated)

Commits

• See full diff in compare view