API Try on Playground Documentation

**Examples**, **tutorials**, and **applications** to help you build with our open-weight [LFMs](https://huggingface.co/LiquidAI) on laptops, mobile, and edge devices. ## Contents - [Desktop Apps](#-desktop-apps) - [Browser Apps](#-browser-apps) - [Mobile Apps](#-mobile-apps) (Android / iOS) - [Fine-Tuning Examples](#-fine-tuning-examples) - [Guides](#-guides) - [Third-Party Apps Powered by LFM](#third-party-apps-powered-by-lfm) - [Community Projects](#-community-projects) - [Technical Deep Dives](#-technical-deep-dives) - [Contributing](#contributing) - [Support](#support) ## 🖥️ Desktop Apps Python and CLI applications for running LFM models on your laptop or desktop machine. | Name | Description | Link | |------|-------------|------| | Invoice Parser | Extract structured data from invoice images using LFM2-VL-3B | [Code](./examples/invoice-parser/README.md) | | Audio Transcription CLI | Real-time audio-to-text transcription using LFM2.5-Audio-1.5B with llama.cpp | [Code](./examples/audio-transcription-cli/) | | Flight Search Assistant | Find and book plane tickets using LFM2.5-1.2B-Thinking with tool calling | [Code](./examples/flight-search-assistant/README.md) | | Audio Car Cockpit | Voice-controlled car cockpit demo combining LFM2.5-Audio-1.5B with LFM2-1.2B-Tool | [Code](./examples/audio-car-cockpit/README.md) | | LocalCowork | On-device AI agent for file ops, security scanning, OCR, and more, powered by LFM2-24B-A2B | [Code](./examples/localcowork/README.md) | | Home Assistant | Local home assistant with tool calling, benchmarking, and fine-tuning pipeline using LFM2-350M and LFM2.5-1.2B | [Code](./examples/home-assistant/README.md) | | Voice Assistant for Mac | On-device voice assistant for Apple Silicon Macs using LFM2.5-Audio-1.5B and the LEAP SDK | [Code](https://github.com/Liquid4All/LeapSDK-Examples/tree/main/leap-ui-demo/macos/LeapVoiceAssistantDemo) | | Vision Assistant for Mac | On-device visual language model chat for Apple Silicon Macs using the LEAP SDK | [Code](https://github.com/Liquid4All/LeapSDK-Examples/tree/main/macOS/LeapVLMExample/LeapVLMExample) | ## 🌐 Browser Apps Zero-install applications running LFM models directly in the browser via WebGPU and ONNX Runtime Web. | Name | Description | Link | |------|-------------|------| | Tool Calling | Run LFM2 entirely in your browser with WebGPU for in-browser tool calling | [Code](https://huggingface.co/spaces/LiquidAI/LFM2-WebGPU/tree/main) \| [Demo](https://huggingface.co/spaces/LiquidAI/LFM2-WebGPU) | | Voice Assistant | Run LFM2.5-Audio-1.5B entirely in your browser for speech recognition, TTS, and conversation | [Code](./examples/audio-webgpu-demo/README.md) \| [Demo](https://huggingface.co/spaces/LiquidAI/LFM2.5-Audio-1.5B-transformers-js) | | Live Video Captioning | Real-time video captioning with LFM2.5-VL-1.6B running in-browser using WebGPU | [Code](./examples/vl-webgpu-demo/README.md) \| [Demo](https://huggingface.co/spaces/LiquidAI/LFM2-VL-WebGPU) | | Chain-of-Thought Reasoning | Run LFM2.5-1.2B-Thinking entirely in your browser with WebGPU for on-device chain-of-thought reasoning | [Code](https://huggingface.co/spaces/LiquidAI/LFM2.5-1.2B-Thinking-WebGPU/tree/main) \| [Demo](https://huggingface.co/spaces/LiquidAI/LFM2.5-1.2B-Thinking-WebGPU) | | Hand & Voice Racer | Browser driving game controlled by hand gestures (MediaPipe) and voice commands (LFM2.5-Audio-1.5B), running fully local | [Code](./examples/hand-voice-racer/README.md) | | Road Decider | Pixel-art survival racer where Liquid's d1 decision model dodges traffic against Jev or a human player | [Code](./examples/road-decider/README.md) | | LEAP Voice Assistant | On-device voice assistant running in the browser via WebAssembly using the LEAP SDK | [Code](https://github.com/Liquid4All/LeapSDK-Examples/tree/main/leap-ui-demo/web) | ## 📱 Mobile Apps Native examples for deploying LFM2 models on iOS and Android using the [LEAP Edge SDK](https://leap.liquid.ai/docs/edge-sdk/overview). Written for Android (Kotlin) and iOS (Swift), the goal of the Edge SDK is to make Small Language Model deployment as easy as calling a cloud LLM API endpoint. ### Android | Name | Description | Link | |------|-------------|------| | LeapChat | Chat app with real-time streaming, persistent history, and modern UI | [Code](https://github.com/Liquid4All/LeapSDK-Examples/tree/main/Android/LeapChat) | | Voice Assistant | Audio input and output with LFM2.5-Audio-1.5B for on-device AI inference | [Code](https://github.com/Liquid4All/LeapSDK-Examples/tree/main/leap-ui-demo/android) | | LeapKoogAgent | Integration with Koog framework for AI agent functionality | [Code](https://github.com/Liquid4All/LeapSDK-Examples/tree/main/Android/LeapKoogAgent) | | SloganApp | Single turn marketing slogan generation with Android Views | [Code](https://github.com/Liquid4All/LeapSDK-Examples/tree/main/Android/SloganApp) | | ShareAI | Website summary generator | [Code](https://github.com/Liquid4All/LeapSDK-Examples/tree/main/Android/ShareAI) | | Recipe Generator | Structured output generation with the LEAP SDK | [Code](https://github.com/Liquid4All/LeapSDK-Examples/tree/main/Android/RecipeGenerator) | | VLM Example | Visual Language Model integration | [Code](https://github.com/Liquid4All/LeapSDK-Examples/tree/main/Android/VLMExample) | ### iOS | Name | Description | Link | |------|-------------|------| | LeapChat | Chat app with real-time streaming, conversation management, and SwiftUI | [Code](https://github.com/Liquid4All/LeapSDK-Examples/tree/main/iOS/LeapChatExample) | | LeapSloganExample | Basic LeapSDK integration for text generation in SwiftUI | [Code](https://github.com/Liquid4All/LeapSDK-Examples/tree/main/iOS/LeapSloganExample) | | Recipe Generator | Structured output generation | [Code](https://github.com/Liquid4All/LeapSDK-Examples/tree/main/iOS/RecipeGenerator) | | Voice Assistant | Audio input and output with LFM2.5-Audio-1.5B for on-device AI inference | [Code](https://github.com/Liquid4All/LeapSDK-Examples/tree/main/leap-ui-demo/ios/LeapVoiceAssistantDemo) | | Vision Assistant | Visual language model chat on iOS using the LEAP SDK | [Code](https://github.com/Liquid4All/LeapSDK-Examples/tree/main/iOS/LeapVLMExample) | | Telco Triage | Edge-first home internet support assistant with on-device LFM routing, local tools, and cloud-assist handoff | [Code](./examples/telco-triage-ios/README.md) | ## 🎯 Fine-Tuning Examples We recommend fine-tuning LFM3 for your specific use case to achieve the best results. Below you'll find Colab notebooks and Python scripts for customizing LFM models with your own data. | Name | Framework | Description | Link | |------|-----------|-------------|------| | **Text Models** | | | | | Supervised Fine-Tuning (SFT) for structured extraction | [TRL](https://github.com/huggingface/trl) ([Docs](https://docs.liquid.ai/lfm/fine-tuning/trl)) | Fine-tune LFM3-1.7B with LoRA to turn messy documents in many languages into JSON that follows a schema | [Notebook](./finetuning/notebooks/lfm3_sft_with_trl.ipynb) [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Liquid4All/cookbook/blob/main/finetuning/notebooks/lfm3_sft_with_trl.ipynb) | | Continued Pre-Training (CPT) for a new language | [TRL](https://github.com/huggingface/trl) ([Docs](https://docs.liquid.ai/lfm/fine-tuning/trl)) | Continue pre-training LFM3-1.7B with LoRA on Turkish Wikipedia, a language it doesn't officially support | [Notebook](./finetuning/notebooks/lfm3_cpt_with_trl.ipynb) [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Liquid4All/cookbook/blob/main/finetuning/notebooks/lfm3_cpt_with_trl.ipynb) | | Direct Preference Optimization (DPO) for grounded answers | [TRL](https://github.com/huggingface/trl) ([Docs](https://docs.liquid.ai/lfm/fine-tuning/trl)) | Teach LFM3-1.7B to answer only from the retrieved documents, and to say so when the answer isn't there | [Notebook](./finetuning/notebooks/lfm3_dpo_with_trl.ipynb) [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Liquid4All/cookbook/blob/main/finetuning/notebooks/lfm3_dpo_with_trl.ipynb) | | Group Relative Policy Optimization (GRPO) for tool calling | [TRL](https://github.com/huggingface/trl) ([Docs](https://docs.liquid.ai/lfm/fine-tuning/trl)) | Train LFM3-1.7B as a phone assistant that picks the right calls among 108 tools, with a verifiable reward and LoRA | [Notebook](./finetuning/notebooks/lfm3_grpo_with_trl.ipynb) [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Liquid4All/cookbook/blob/main/finetuning/notebooks/lfm3_grpo_with_trl.ipynb) | | Continued Pre-Training (CPT) for text completion | [Unsloth](https://github.com/unslothai/unsloth) ([Docs](https://docs.liquid.ai/lfm/fine-tuning/unsloth)) | Teach models domain-specific knowledge and creative writing styles | [Notebook](./finetuning/notebooks/cpt_text_completion_with_unsloth.ipynb) [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Liquid4All/cookbook/blob/main/finetuning/notebooks/cpt_text_completion_with_unsloth.ipynb) | | Continued Pre-Training (CPT) for translation | [Unsloth](https://github.com/unslothai/unsloth) ([Docs](https://docs.liquid.ai/lfm/fine-tuning/unsloth)) | Adapt models to specific languages or translation domains using domain data | [Notebook](./finetuning/notebooks/cpt_translation_with_unsloth.ipynb) [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Liquid4All/cookbook/blob/main/finetuning/notebooks/cpt_translation_with_unsloth.ipynb) | | Supervised Fine-Tuning (SFT) | [Unsloth](https://github.com/unslothai/unsloth) ([Docs](https://docs.liquid.ai/lfm/fine-tuning/unsloth)) | Memory-efficient SFT using Unsloth with LoRA for 2x faster training | [Notebook](./finetuning/notebooks/sft_with_unsloth.ipynb) [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Liquid4All/cookbook/blob/main/finetuning/notebooks/sft_with_unsloth.ipynb) | | Supervised Fine-Tuning (SFT) | [TRL](https://github.com/huggingface/trl) ([Docs](https://docs.liquid.ai/lfm/fine-tuning/trl)) | Supervised fine-tuning using Hugging Face TRL library with parameter-efficient LoRA | [Notebook](./finetuning/notebooks/sft_with_trl.ipynb) [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Liquid4All/cookbook/blob/main/finetuning/notebooks/sft_with_trl.ipynb) | | Supervised Fine-Tuning (SFT) for Mixture-of-Experts | [Halo](https://github.com/whitecircle/halo) ([Docs](https://whitecircle.com/research/halo)) | Supervised fine-tuning of LFM2.5-8B-A1B with two-way expert parallelism using Halo | [Notebook](./finetuning/notebooks/sft_moe_with_halo.ipynb) | | Direct Preference Optimization (DPO) | [TRL](https://github.com/huggingface/trl) ([Docs](https://docs.liquid.ai/lfm/fine-tuning/trl)) | Align models with human preferences using direct preference optimization with LoRA | [Notebook](./finetuning/notebooks/dpo_with_trl.ipynb) [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Liquid4All/cookbook/blob/main/finetuning/notebooks/dpo_with_trl.ipynb) | | Antidoom Training (FTPO) | [TRL](https://github.com/huggingface/trl) ([Docs](https://huggingface.co/docs/trl/main/en/customization#change-the-training-objective)) | Eliminate doom loops with final token preference optimization (FTPO) using the [Antidoom](https://github.com/Liquid4All/antidoom) pipeline | [Notebook](./finetuning/notebooks/antidoom_with_trl.ipynb) [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Liquid4All/cookbook/blob/main/finetuning/notebooks/antidoom_with_trl.ipynb) | | Group Relative Policy Optimization (GRPO) | [TRL](https://github.com/huggingface/trl) ([Docs](https://docs.liquid.ai/lfm/fine-tuning/trl)) | Train reasoning models using Group Relative Policy Optimization with rule-based rewards | [Notebook](./finetuning/notebooks/grpo_for_verifiable_tasks.ipynb) [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Liquid4All/cookbook/blob/main/finetuning/notebooks/grpo_for_verifiable_tasks.ipynb) | | Group Relative Policy Optimization (GRPO) | [TRL](https://github.com/huggingface/trl) ([Docs](https://docs.liquid.ai/lfm/fine-tuning/trl)) | Train a 350M model for better structured output compliance | [Notebook](./finetuning/notebooks/grpo_with_trl_ifstruct.ipynb) [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Liquid4All/cookbook/blob/main/finetuning/notebooks/grpo_with_trl_ifstruct.ipynb) | | Group Relative Policy Optimization (GRPO) | [Unsloth](https://github.com/unslothai/unsloth) ([Docs](https://docs.liquid.ai/lfm/fine-tuning/unsloth)) | Train reasoning models using Group Relative Policy Optimization for verifiable tasks | [Notebook](./finetuning/notebooks/grpo_with_unsloth.ipynb) [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Liquid4All/cookbook/blob/main/finetuning/notebooks/grpo_with_unsloth.ipynb) | | **Decision Models** | | | | | Supervised Fine-Tuning (SFT) | [TRL](https://github.com/huggingface/trl) ([Docs](https://docs.liquid.ai/lfm/fine-tuning/trl)) | Learn how to fine-tune a decision model (d1-3B) on your own labels with LoRA, using intent classification on Banking77 as the example | [Notebook](./finetuning/notebooks/sft_for_decision_models_with_trl.ipynb) [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Liquid4All/cookbook/blob/main/finetuning/notebooks/sft_for_decision_models_with_trl.ipynb) | | Supervised Fine-Tuning (SFT) | [Unsloth](https://github.com/unslothai/unsloth) ([Docs](https://docs.liquid.ai/lfm/fine-tuning/unsloth)) | Learn how to fine-tune a decision model (d1-3B) on your own labels with Unsloth and LoRA, using intent classification on Banking77 as the example | [Notebook](./finetuning/notebooks/sft_for_decision_models_with_unsloth.ipynb) [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Liquid4All/cookbook/blob/main/finetuning/notebooks/sft_for_decision_models_with_unsloth.ipynb) | | **Vision Models** | | | | | Supervised Fine-Tuning (SFT) for receipt extraction | [TRL](https://github.com/huggingface/trl) ([Docs](https://docs.liquid.ai/lfm/fine-tuning/trl)) | Fine-tune LFM3-VL-2B with LoRA to read photos of receipts into JSON that follows a schema | [Notebook](./finetuning/notebooks/lfm3_vl_sft_with_trl.ipynb) [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Liquid4All/cookbook/blob/main/finetuning/notebooks/lfm3_vl_sft_with_trl.ipynb) | | Group Relative Policy Optimization (GRPO) for formula transcription | [TRL](https://github.com/huggingface/trl) ([Docs](https://docs.liquid.ai/lfm/fine-tuning/trl)) | Train LFM3-VL-2B with LoRA to transcribe pictures of formulas into LaTeX, with a format reward and a LaTeX similarity reward | [Notebook](./finetuning/notebooks/lfm3_vl_grpo_with_trl.ipynb) [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Liquid4All/cookbook/blob/main/finetuning/notebooks/lfm3_vl_grpo_with_trl.ipynb) | | Supervised Fine-Tuning (SFT) | [Unsloth](https://github.com/unslothai/unsloth) ([Docs](https://docs.liquid.ai/lfm/fine-tuning/unsloth)) | Supervised fine-tuning for LFM2-VL models on custom image-text datasets | [Notebook](./finetuning/notebooks/sft_for_vision_language_model_with_unsloth.ipynb) [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Liquid4All/cookbook/blob/main/finetuning/notebooks/sft_for_vision_language_model_with_unsloth.ipynb) | | Supervised Fine-Tuning (SFT) | [TRL](https://github.com/huggingface/trl) ([Docs](https://docs.liquid.ai/lfm/fine-tuning/trl)) | Supervised fine-tuning for LFM2.5-VL models using TRL with parameter-efficient LoRA | [Notebook](./finetuning/notebooks/sft_for_vision_language_model_with_trl.ipynb) [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Liquid4All/cookbook/blob/main/finetuning/notebooks/sft_for_vision_language_model_with_trl.ipynb) | | Supervised Fine-Tuning (SFT) | [Halo](https://github.com/whitecircle/halo) ([Docs](https://whitecircle.com/research/halo)) | Supervised fine-tuning for LFM2.5-VL-3B on CPPE-5 for object detection | [Notebook](./finetuning/notebooks/sft_for_vision_language_model_with_halo.ipynb) | | Wind Turbine Blade Inspection | [TRL](https://github.com/huggingface/trl) ([Docs](https://docs.liquid.ai/lfm/fine-tuning/trl)) | Fine-tune LFM3-VL-2B with SFT and GRPO to find and classify surface defects on wind turbine blade photos, and compare it with frontier models | [Code](./examples/wind-turbine-inspection/README.md) | | Satellite VLM Fine-Tuning | [leap-finetune](https://github.com/Liquid4All/leap-finetune/) | Fine-tune LFM2.5-VL-450M on satellite imagery for VQA, grounding, and captioning using Modal | [Code](./examples/satellite-vlm/README.md) | | Wildfire Prevention | [leap-finetune](https://github.com/Liquid4All/leap-finetune/) | Build a wildfire risk detection system using LFM2.5-VL-450M and Sentinel-2 satellite imagery, with fine-tuning and on-device inference | [Code](./examples/wildfire-prevention/README.md) | | **Audio Models** | | | | | Supervised Fine-Tuning (SFT) for voice commands | [HF Transformers](https://github.com/huggingface/transformers) | Fine-tune LFM3-Audio-2B with LoRA to map spoken smart-home commands directly to function calls | [Notebook](./finetuning/notebooks/lfm3_audio_sft_with_transformers.ipynb) [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Liquid4All/cookbook/blob/main/finetuning/notebooks/lfm3_audio_sft_with_transformers.ipynb) | | Supervised Fine-Tuning (SFT) for a new voice | [HF Transformers](https://github.com/huggingface/transformers) | Teach LFM3-Audio-2B a new, consistent voice by training its speech output on LJSpeech | [Notebook](./finetuning/notebooks/lfm3_audio_tts_sft_with_transformers.ipynb) [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Liquid4All/cookbook/blob/main/finetuning/notebooks/lfm3_audio_tts_sft_with_transformers.ipynb) | | LFM2.5-Audio Fine-Tuning | [liquid-audio](https://pypi.org/project/liquid-audio/) | Fine-tune LFM2.5-Audio-1.5B on the OHF-Voice dataset to map speech directly to Home Assistant function calls, with on-device GGUF inference | [Code](./examples/voice-assistant/README.md) | | **Omni Models** | | | | | Supervised Fine-Tuning (SFT) for voice commands | [HF Transformers](https://github.com/huggingface/transformers) | Fine-tune LFM3-Omni-2B with LoRA to map spoken smart-home commands directly to function calls | [Notebook](./finetuning/notebooks/lfm3_omni_sft_with_transformers.ipynb) [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Liquid4All/cookbook/blob/main/finetuning/notebooks/lfm3_omni_sft_with_transformers.ipynb) | | Supervised Fine-Tuning (SFT) for spoken image Q&A | [HF Transformers](https://github.com/huggingface/transformers) | Fine-tune LFM3-Omni-2B to answer spoken questions about medical images | [Notebook](./finetuning/notebooks/lfm3_omni_vqa_sft_with_transformers.ipynb) [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Liquid4All/cookbook/blob/main/finetuning/notebooks/lfm3_omni_vqa_sft_with_transformers.ipynb) | | **Encoder Models** | | | | | Document Classification | [HF Transformers](https://github.com/huggingface/transformers) | Fine-tune LFM2.5-Encoder for custom long-document multi-label classification | [Code](./examples/lfm-encoder-classification/) | ## 📓 Guides Notebooks walking through the capabilities of specific LFM models with Hugging Face Transformers. | Name | Description | Link | |------|-------------|------| | Vision Capabilities | Image description, visual question answering, multi-image comparison, text reading, object localization, and tool calling with LFM2.5-VL | [Notebook](./guides/lfm25_vl_image_understanding.ipynb) [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Liquid4All/cookbook/blob/main/guides/lfm25_vl_image_understanding.ipynb) | | Longevity Models | Interpret clinical, DNA methylation, transcriptomic, proteomic, and genetic aging data with LFM2-1.2B-Longevity and LFM2-2.6B-Longevity | [Notebook](./guides/lfm2_longevity.ipynb) [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/Liquid4All/cookbook/blob/main/guides/lfm2_longevity.ipynb) | ## Third-Party Apps Powered by LFM Production and open-source applications that support LFM models as an inference backend, among other providers. | Name | Description | Link | |------|-------------|------| | DeepCamera | Open-source AI camera system for local vision intelligence with facial recognition, person re-ID, and edge deployment on Jetson and Raspberry Pi | [Code](https://github.com/SharpAI/DeepCamera) | | Osaurus | Native macOS AI harness for managing agents, memory, tools, and identity locally, with support for LFM models via MLX on Apple Silicon | [Code](https://github.com/osaurus-ai/osaurus) | | SelfLink | Native iOS privacy-first journaling app | [Website](https://www.realityplay.io/selflink) | ## 🌟 Community Projects Open-source projects built by the community showcasing LFMs with real use cases. ### Fine-tuning | Name | Description | Link | |------|-------------|------| | LFM2.5 Mobile Actions | LoRA fine-tuned LFM2.5-1.2B that translates natural language into Android OS function calls for on-device mobile action recognition | [Code](https://github.com/Mandark-droid/LFM2.5-1.2B-Instruct-mobile-actions) | | Food Images Fine-tuning | Fine-tune LFM models on food image datasets | [Code](https://github.com/benitomartin/food-images-finetuning) | | LFM2-KoEn-Tuning | Fine-tuned LFM2 1.2B for Korean-English translation | [Code](https://github.com/gyunggyung/LFM2-KoEn-Tuning) | | SFT + DPO Fine-tuning | Teaching a 1.2B model to be a grumpy Italian chef: SFT + DPO fine-tuning with Unsloth | [Code](https://github.com/benitomartin/grumpy-chef-finetuning-dpo) | | LFM2-2.6B Mr. Tic Tac Toe | Fine-tune LFM2-2.6B with reinforcement learning to play tic-tac-toe | [Code](https://github.com/anakin87/llm-rl-environments-lil-course) | ### Deployment | Name | Description | Link | |------|-------------|------| | LFM-2.5 Thinking on Web | Run LFM2.5-1.2B reasoning model locally in the browser via WebGPU and Transformers.js | [Code](https://github.com/sitammeur/lfm2.5-thinking-web) | | LFM-2.5 JP on Web | Run LFM2.5-1.2B Japanese model locally in the browser via WebGPU and Transformers.js | [Code](https://github.com/sitammeur/lfm2.5-jp-web) | | barq-web-rag | Browser-based RAG app for document Q&A with LFM2.5-1.2B-Thinking running fully local via WebGPU | [Code](https://github.com/YASSERRMD/barq-web-rag) | | Tauri Plugin LEAP AI | Tauri plugin to integrate LEAP and Liquid LFMs into desktop and mobile apps | [Crate](https://crates.io/crates/tauri-plugin-leap-ai) | | Chat with LEAP SDK | LEAP SDK integration for React Native | [Code](https://github.com/glody007/expo-leap-sdk) | ### End-to-End Projects | Name | Description | Link | |------|-------------|------| | Image Classification on Edge | Fine-tune and deploy a local VLM for fast, accurate image classification on edge devices | [Code](https://github.com/Paulescu/image-classification-with-local-vlms) | | Chess Game with Small LMs | Fine-tune and deploy a small language model to play chess | [Code](https://github.com/Paulescu/chess-game) | | Private Doc Q&A | On-device document Q&A with RAG and voice input | [Code](https://github.com/chintan-projects/private-doc-qa) | | Photo Triage Agent | Private photo library cleanup using LFM vision model | [Code](https://github.com/chintan-projects/photo-triage-agent) | | Tiny-MoA | Mixture of agents on CPU with LFM2.5 Brain (1.2B) | [Code](https://github.com/gyunggyung/Tiny-MoA) | | LFM-Scholar | Automated literature review agent for finding and citing papers | [Code](https://github.com/gyunggyung/LFM-Scholar) | | Private Summarizer | 100% local text summarization with multi-language support | [Code](https://github.com/Private-Intelligence/private_summarizer) | | TranslatorLens | Offline translation camera for real-time text translation | [Code](https://github.com/linmx0130/TranslatorLens) | | Meeting Intelligence CLI | CLI tool for meeting transcription and analysis | [Code](https://github.com/chintan-projects/meeting-prompter) | | grosme | CLI grocery assistant that finds Walmart product matches using an LFM-2.5 tool-calling agent via Ollama | [Code](https://github.com/earl562/grosme) | | Discord Moderator | Use LFM2.5-1.2B to screen messages for suspicious content | [Code](https://github.com/badluma/liquid-mod) | | BookMind | Offline RAG study assistant for asking questions and generating exercises from PDF textbooks, powered by LFM2-2.6B | [Code](https://github.com/Ksirailway-base/BookMind) | | Liquid-CLI | Fine-tune and run Liquid AI's LFM2-8B-A1B as a local terminal agent | [Code](https://github.com/gyunggyung/Liquid-CLI) | | LFM Podcast Studio | Turn any PDF into a two-host podcast episode locally using LFM2.5-Audio TTS via llama.cpp | [Code](https://github.com/nikhilprasanth/LFM-Podcast-Studio) | | GalamseyWatch | Two-layer agentic Earth observation over Sentinel-2 imagery for illegal mining detection in Ghana, powered by LFM2.5-VL-450M and LFM2-2.6B | [Code](https://github.com/samadon1/GalamseyWatch) | | Liquid Lens | Detects water anomalies in orbit by combining fast spectral filtering with conditional LFM2.5-VL visual insights on the SimSat orbital simulator | [Code](https://github.com/LiquidLensSystems/water-vlm) | ## 🕐 Technical Deep Dives Recorded sessions (~60 minutes) covering advanced topics and hands-on implementations. | Date | Topic | Link | |------|-------|------| | 2025-11-06 | Fine-tuning LFM2-VL for image classification | [Video](https://www.youtube.com/watch?v=00IK9apncCg) | | 2025-11-27 | Building a 100% local Audio-to-Speech CLI with LFM2-Audio | [Video](https://www.youtube.com/watch?v=yeu077gPmCA) | | 2025-12-26 | Fine-tuning LFM2-350M for browser control with GRPO and OpenEnv | [Video](https://www.youtube.com/watch?v=gKQ08yee3Lw) | | 2026-01-22 | Local video-captioning with LFM2.5-VL-1.6B and WebGPU | [Video](https://www.youtube.com/watch?v=xsWARHFoA3E) | | 2026-03-05 | Build your own local AI coding assistant | [Video](https://www.youtube.com/watch?v=6JEm1IxcxEw) | | 2026-04-22 | Let's build a wildfire detection system with a Vision Language Model and satellite images | [Video](https://www.youtube.com/watch?v=LOIDYl5fdb8) | | 2026-06-12 | Building on-device voice agents | [Video](https://www.youtube.com/watch?v=0YQquPG-9R8) | Join the next session! Head to the `#live-events` channel on [Discord](https://discord.com/invite/liquid-ai). ## Contributing We welcome contributions! Open a PR with a link to your project GitHub repo in the Community Projects section. [![Contributors](https://contrib.rocks/image?repo=Liquid4All/cookbook)](https://github.com/Liquid4All/cookbook/graphs/contributors) ## Support - 📖 [Liquid AI Documentation](https://docs.liquid.ai/) - 💬 [Join our community on Discord](https://discord.com/invite/liquid-ai)