Llamatik - Local AI Chatbot
100+
Downloads
Everyone
info
About this app
Llamatik — Local AI Chatbot Powered by On-Device Models
Llamatik is an offline AI chatbot designed to showcase the capabilities of the Llamatik Kotlin Multiplatform library. All AI text generation happens directly on your device, using lightweight, efficient open-source models. No internet connection is required for chat, and no conversations are ever sent to any server.
Whether you want to summarize text, create ideas, draft content, or simply explore what a fully on-device LLM can do, Llamatik makes it fast, private, and reliable.
⸻
🔒 100% On-Device AI
All text generation runs locally.
Your prompts, responses, and model interactions never leave your device.
⸻
⚡ Fast and Lightweight
Llamatik supports small and efficient models such as:
• Gemma 3 270M
• SmolVLM 256M / 500M
• Phi-1.5
• Qwen 2.5 5B (quantized)
• Llama 3.2 1B (quantized)
These models run smoothly on modern Android devices without requiring cloud compute.
⸻
📦 Built-In Model Downloader
Easily download, install, and switch between supported models.
Choose the model that best fits your device performance and your use case.
⸻
🎨 Clean & Modern Interface
A friendly interface makes chatting intuitive:
• Smart suggestions
• Quick prompts
• Model selector
• Beautiful onboarding
• Minimalist design
⸻
🔐 Privacy First
Llamatik does not collect, store, or transmit chat content.
Your data remains on your device at all times.
⸻
📱 Experimental Showcase
Llamatik is an experimental demo created to help developers and curious users experience the capabilities of the Llamatik library on mobile devices.
⸻
💡 Notes
• Requires downloading at least one model to begin chatting.
• Internet is only used to download models or for analytics/crash reporting (Firebase).
• Chat content is fully offline.
Llamatik is an offline AI chatbot designed to showcase the capabilities of the Llamatik Kotlin Multiplatform library. All AI text generation happens directly on your device, using lightweight, efficient open-source models. No internet connection is required for chat, and no conversations are ever sent to any server.
Whether you want to summarize text, create ideas, draft content, or simply explore what a fully on-device LLM can do, Llamatik makes it fast, private, and reliable.
⸻
🔒 100% On-Device AI
All text generation runs locally.
Your prompts, responses, and model interactions never leave your device.
⸻
⚡ Fast and Lightweight
Llamatik supports small and efficient models such as:
• Gemma 3 270M
• SmolVLM 256M / 500M
• Phi-1.5
• Qwen 2.5 5B (quantized)
• Llama 3.2 1B (quantized)
These models run smoothly on modern Android devices without requiring cloud compute.
⸻
📦 Built-In Model Downloader
Easily download, install, and switch between supported models.
Choose the model that best fits your device performance and your use case.
⸻
🎨 Clean & Modern Interface
A friendly interface makes chatting intuitive:
• Smart suggestions
• Quick prompts
• Model selector
• Beautiful onboarding
• Minimalist design
⸻
🔐 Privacy First
Llamatik does not collect, store, or transmit chat content.
Your data remains on your device at all times.
⸻
📱 Experimental Showcase
Llamatik is an experimental demo created to help developers and curious users experience the capabilities of the Llamatik library on mobile devices.
⸻
💡 Notes
• Requires downloading at least one model to begin chatting.
• Internet is only used to download models or for analytics/crash reporting (Firebase).
• Chat content is fully offline.
Updated on
Safety starts with understanding how developers collect and share your data. Data privacy and security practices may vary based on your use, region, and age. The developer provided this information and may update it over time.
No data shared with third parties
Learn more about how developers declare sharing
This app may collect these data types
App info and performance
Data is encrypted in transit
Data can’t be deleted
What’s new
✨ New in this update!
• New Generation Settings panel to fine-tune AI responses
• Customize temperature, max tokens, top-p, top-k, and repeat penalty
• Improved generation with per-model prompt templates
• Correct model handling for Gemma 3 with the chat runner
• Android settings updates and UI fixes (slider values now display correctly)
More control, better outputs, smoother local AI. 🚀
• New Generation Settings panel to fine-tune AI responses
• Customize temperature, max tokens, top-p, top-k, and repeat penalty
• Improved generation with per-model prompt templates
• Correct model handling for Gemma 3 with the chat runner
• Android settings updates and UI fixes (slider values now display correctly)
More control, better outputs, smoother local AI. 🚀
App support
About the developer
Ferran Pons Sanchez
support@multiplatformkickstarter.com
Spain
