Llamatik and ToolNeuron
These are **competitors** offering overlapping on-device AI capabilities (LLM inference via llama.cpp, speech-to-text, image generation) for mobile/edge platforms, with ToolNeuron providing broader feature coverage (VLM, RAG, tool calling) while Llamatik emphasizes cross-platform Kotlin Multiplatform support.
About Llamatik
ferranpons/Llamatik
True on-device AI for Kotlin Multiplatform (Android, iOS, Desktop, JVM, WASM). LLM, Speech-to-Text and Image Generation — powered by llama.cpp, whisper.cpp and stable-diffusion.cpp.
Based on the README, here's the technical summary: Wraps **llama.cpp**, **whisper.cpp**, and **stable-diffusion.cpp** behind a unified Kotlin API with **schema-constrained JSON generation** and streaming support for LLMs. Native C++ integration via Kotlin/Native (iOS frameworks) and JNI (Android/Desktop) enables true offline inference without Python dependencies. Supports hybrid operation—seamlessly switch between on-device and HTTP-based remote inference with identical API calls, backed by an optional `llamatik-backend` server.
About ToolNeuron
Siddhesh2377/ToolNeuron
On-device AI for Android — LLM chat (GGUF/llama.cpp), vision models (VLM), image generation (Stable Diffusion), tool calling, AI personas, RAG knowledge packs, TTS/STT. Fully offline, zero subscriptions, open-source.
Implements a hybrid RAG pipeline combining FTS4 BM25, vector search, and Reciprocal Rank Fusion across PDFs, docs, and chat history, with grammar-constrained JSON for function calling. Built on llama.cpp for inference, LocalDream for Stable Diffusion, and a custom Unified Memory System (UMS) for efficient binary storage, with AES-256-GCM encryption backed by Android KeyStore and hardware TEE where available. Includes 7 callable plugins (web search, file ops, calculator) and adaptive hardware tuning that auto-detects CPU topology to optimize thread allocation and context windows.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work