Skip to content
pubscout

Compare

The strongest value in each row is marked.

Attributellama_flutter_android

Run llama.cpp GGUF models on Android via a Flutter plugin

pub.devRemove
nobodywho

Run local LLMs, vision, embeddings, RAG, tool calling, STT, and TTS on mobile and desktop.

pub.devRemove
ScoresHealth5480 (best)
Maintenance5092 (best)
Quality96 (best)85
Popularity3542 (best)
Community1180 (best)
Compatibility75100 (best)
UsageDownloads (30 days)4k (best)1.3k
Likes933 (best)
Pub points150 / 160160 / 160 (best)
GitHub stars261.7k (best)
MaintenanceLatest release0.2.6, 4 months ago6.0.0, 3 weeks ago (best)
Major versions in 12 months0 (best)6
Contributors220 (best)
PublisherUnverifiednobodywho.ooo
FitPlatformsAndroid, iOSAndroid, iOS, Windows, macOS, Linux
WASM-readyYesYes
KindPlugin (native code)Plugin (native code)
Code generationNoYes
Direct dependencies1 (best)4
LicenseMITEUPL-1.2
Setup effortMediumLow
JudgementStrengths
  • Android‑only GPU acceleration via Vulkan
  • Fine‑grained control of sampling, penalties, and templates
  • Stable streaming with guaranteed onDone termination
  • No external network dependency
  • Runs completely offline with no API keys required.
  • Supports text, vision, audio, embeddings, RAG, and tool calling in one library.
  • Hardware acceleration via Metal and Vulkan.
Weaknesses
  • Only works on Android (iOS support missing)
  • Requires NDK r27+ and ARM64 ABI
  • No built‑in background service for long‑running inference
  • Setup requires proguard rules and native library handling
  • Large model sizes significantly increase app download or memory footprints.
  • Performance depends heavily on the user's device specs and available RAM.
Warnings
  • Native setup
  • File-level copyleft
  • Native setup
llama_flutter_android vs nobodywho: which Flutter package should you use? | pubscout