Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/afondiel/awesome-smol
An awesome list of "small but mighty" models and resources.
https://github.com/afondiel/awesome-smol
List: awesome-smol
ai ai-frontier audio awesome-smol computer-vision deep-learning edge-ai edge-computing edge-devices edge-ml embedded-ai generative-ai lightweight-models llms multimodality on-device-ai smls smol-models vision-models
Last synced: 6 days ago
JSON representation
An awesome list of "small but mighty" models and resources.
- Host: GitHub
- URL: https://github.com/afondiel/awesome-smol
- Owner: afondiel
- License: apache-2.0
- Created: 2024-11-23T15:08:58.000Z (3 months ago)
- Default Branch: main
- Last Pushed: 2024-12-20T14:56:43.000Z (2 months ago)
- Last Synced: 2025-02-05T11:02:31.452Z (16 days ago)
- Topics: ai, ai-frontier, audio, awesome-smol, computer-vision, deep-learning, edge-ai, edge-computing, edge-devices, edge-ml, embedded-ai, generative-ai, lightweight-models, llms, multimodality, on-device-ai, smls, smol-models, vision-models
- Homepage:
- Size: 142 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- Contributing: CONTRIBUTING.md
- License: LICENSE
Awesome Lists containing this project
- ultimate-awesome - awesome-smol - An awesome list of "small but mighty" models and resources. (Other Lists / Julia Lists)
README
# Awesome Smol
[](https://awesome.re) [](./CONTRIBUTING.md) [](./LICENSE)
**Awesome Smol** is a curated list of small, lightweight AI models, tools, and resources for domains like **language**, **audio**, **vision**, and **multimodal tasks**. These models are designed for edge devices, resource-constrained environments, and rapid prototyping.
Inspiration: [awesome-tensorflow-lite](https://github.com/margaretmz/awesome-tensorflow-lite).
### **What are Smol Models?**
(Source: [Link](https://x.com/LoubnaBenAllal1/status/1852055587275895188))
Smol models are AI models optimized for efficiency, offering:
- **Lightweight Design:** Minimal memory usage.
- **Fast Inference:** High performance on limited hardware.
- **Accessibility:** Ideal for mobile, IoT, and edge deployment.
- **Versatility:** Support across text, audio, vision, and more.
### **Goal**
- Highlight community-driven advancements in lightweight AI.
- Provide a central hub for small model resources, tools, and benchmarks.
- Promote faster adoption for real-world applications.
### **Contribute to the List**
Your contributions are warmly welcome! Submit a pull request (PR) following the [contribution guidelines](CONTRIBUTING.md).
## **Table of Contents**
- [Language Models](#language-models)
- [Audio Models](#audio-models)
- [Vision Models](#vision-models)
- [Multimodal Models](#multimodal-models)
- [Pretrained Models Hub](#pretrained-models-hub)
- [Other Insightful Lists](#other-insightful-lists)
- [Tools and Frameworks](#tools-and-frameworks)
- [AI Frontier solution at the Edge](#ai-frontier-solution-at-the-edge)
- [AI News & Announcements](#ai-news--announcements)
- [Resources](#resources)
## **Language Models**
| **Model** | **Task** | **Platform** | **References** |
|----------------------------|---------------------|--------------------|-----------------------------------------------------------------|
| SmolLM | General NLP | Edge, Desktop | [Hugging Face](https://huggingface.co/smolai/smollm-1.7b) |
| Zamba2-7B | Text Understanding | Mobile, Edge | [Hugging Face](https://huggingface.co/smolai/zamba2-7b) |
| EuroLLM-1.7B | Multilingual NLP | Mobile, IoT | [Hugging Face](https://huggingface.co/smolai/eurollm-1.7b) |
| Mistral-Small-Instruct-2409| Instruction Tasks | Edge, Mobile | [Hugging Face](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) |
| Ministral-8B-Instruct-2410| Instruction Tasks | Edge, Desktop | [Hugging Face](https://huggingface.co/microsoft/phi-2) |
| TinyLlama | Conversational AI | Edge, IoT | [Hugging Face](https://huggingface.co/TinyLlama/TinyLlama-1.1B-Chat-v1.0) |
| Phi-3 | Text Generation | Mobile, Edge | [Hugging Face](https://huggingface.co/microsoft/phi-2) |
| Gemma 2 | Multilingual NLP | Mobile, Desktop | [Hugging Face](https://huggingface.co/google/gemma-2b) |
## **Audio Models**
| **Model** | **Task** | **Platform** | **References** |
|--------------------|--------------------|----------------------|------------------------------------------------------------------|
| Whisper Small | Speech Recognition | Edge, Desktop | [Hugging Face](https://huggingface.co/openai/whisper-small) |
| Audio-Mamba (AuM) | Audio Processing | Edge, IoT | [GitHub](https://github.com/kaistmm/Audio-Mamba-AuM) |
| MusicGen | Music Generation | Edge, Desktop | [GitHub](https://github.com/facebook/MusicGen) |
| FastSpeech2 Small | Text-to-Speech | Mobile, Edge | [Hugging Face](https://huggingface.co/facebook/fastspeech2-small) |
| HiFi-GAN Mini | Audio Enhancement | Mobile, IoT | [GitHub](https://github.com/jik876/hifi-gan) |
| MatchboxNet Small | Keyword Spotting | Edge, IoT | [Hugging Face](https://huggingface.co/speechbrain/matchboxnet-small) |
## **Vision Models**
| **Model** | **Task** | **Platform** | **References** |
|---------------------------|----------------------|----------------------|------------------------------------------------------------------|
| MobileNet V3 Small | Image Classification | Mobile, Edge | [TensorFlow](https://www.tensorflow.org/lite/models/imagenet) |
| EfficientNet-Lite Small | Image Classification | Mobile, IoT | [GitHub](https://github.com/google/automl/tree/master/efficientnet) |
| YOLOv5 Nano | Object Detection | Edge, IoT | [GitHub](https://github.com/ultralytics/yolov5) |
| DeepLab Lite Small | Image Segmentation | Mobile, Edge | [GitHub](https://github.com/tensorflow/models/tree/master/research/deeplab) |
| MobileUNet | Image Segmentation | Mobile, IoT | [GitHub](https://github.com/zhixuhao/unet) |
| Vision Transformer Small | Vision Tasks | Mobile, Edge | [Hugging Face](https://huggingface.co/models?filter=vit) |
## **Multimodal Models**
| **Model** | **Task** | **Platform** | **References** |
|-------------------|-----------------------|----------------------|------------------------------------------------------------------|
| Mini-DALL-E | Text-to-Image | Mobile, Desktop | [GitHub](https://github.com/borisdayma/dalle-mini) |
| TinyCLIP | Vision-Language | Edge, IoT | [Hugging Face](https://huggingface.co/openai/clip) |
| Mini-ALIGN | Vision-Language | Mobile, Edge | [GitHub](https://github.com/google-research/align) |
## **Pretrained Models Hub**
Pre-trained lightweight models ready for deployment:
- [Hugging Face Pretrained Smol Models](https://huggingface.co/models?sort=trending&search=smol): Ready-to-deploy smol models with associated datasets, and demo apps (Spaces).
- [Model Zoo Models Categories](https://modelzoo.co/categories): Open source deep learning code and pretrained models.
- [Kaggle Pre-trained Models](https://www.kaggle.com/models): Use and download pre-trained models for your machine learning projects.
- [Tensorflow Hub](https://www.tensorflow.org/hub): A repository of trained machine learning models.
- [Pytorch Hub](https://pytorch.org/hub/): Discover and publish models to a pre-trained model repository designed for research exploration.
## **Other Insightful Lists**
- [edge-ai - @crespum](https://github.com/crespum/edge-ai)
- [awesome-tensorflow-lite - @margaretmz](https://github.com/margaretmz/awesome-tensorflow-lite)
- [Smol Vision - @merveenoyan](https://github.com/merveenoyan/smol-vision)
- [Edge AI Model Zoo - @afondiel](https://github.com/afondiel/EdgeAI-Model-Zoo)
## **Tools and Frameworks**
- [LiteRT - formerly TensorFlow Lite](https://ai.google.dev/edge/litert): Lightweight model deployment for Android.
- [CoreML](https://developer.apple.com/documentation/coreml): Apple’s ML framework for iOS.
- [ExecuTorch](https://pytorch.org/executorch-overview): Pytorch/Meta end-to-end solution for enabling on-device inference.
- [ONNX Runtime](https://onnxruntime.ai/): Efficient inference engine.
- [OpenVINO](https://github.com/openvinotoolkit/openvino): Toolkit for optimizing and deploying deep learning models.
## **AI Frontier solution at the Edge**
- [Google AI Edge](https://ai.google.dev/edge)
- [AWS IoT for the Edge](https://aws.amazon.com/iot/solutions/iot-edge/)
- [Azure IoT Edge - Build the intelligent edge](https://azure.microsoft.com/en-us/products/iot-edge/)
- [Qualcomm On-Device AI Solutions](https://www.qualcomm.com/edgeofpossible)
- [Meta - Pytorch Edge](https://pytorch.org/edge)
- [NVIDIA TensorRT](https://developer.nvidia.com/tensorrt-getting-started)
- [Edge Impulse](https://docs.edgeimpulse.com/docs)
- [Edge AI + Vision Alliance](https://www.edge-ai-vision.com/)
- [Edge AI Foundation](https://www.edgeaifoundation.org/)
## **AI News & Announcements**
- **
[2024/11/26]** [SmolVLM - small yet mighty Vision Language Model](https://huggingface.co/blog/smolvlm)
- **[2024/09/25]** [Meta - Llama 3.2: Revolutionizing edge AI and vision with open, customizable models](https://ai.meta.com/blog/llama-3-2-connect-2024-vision-edge-mobile-devices/)
- **[2024/07/16]** [SmolLM - blazingly fast and remarkably powerful](https://huggingface.co/blog/smollm)
- **[2024/04/23]** [Introducing Phi-3: Redefining what’s possible with SLMs](https://azure.microsoft.com/en-us/blog/introducing-phi-3-redefining-whats-possible-with-slms/)
## **Resources**
### Practical Guides
- [Edge-AI core concepts for all levels](https://github.com/afondiel/computer-science-notebook/tree/master/core/systems/edge-computing/edge-ai/concepts/)
- [ML Optimization Resources](https://github.com/afondiel/computer-science-notebook/tree/master/core/ai-ml/ml-notes/model-optimization)
### Crash-Courses & Moocs
- [Introduction to On-Device AI - Qualcomm](https://github.com/afondiel/Introduction-to-On-Device-AI-DLAI)
- [Introduction to edge AI - Edge Impulse](https://docs.edgeimpulse.com/docs/concepts/edge-ai/intro-to-edge-ai)
### **Blogs**
- [Why Small Language Models (SLMs) Are The Next Big Thing In AI - Forbes (2024/11/25)](https://www.forbes.com/sites/deandebiase/2024/11/25/why-small-language-models-are-the-next-big-thing-in-ai/)
- [Optimizing Generative AI for Edge Devices](https://www.qualcomm.com/news/onq/2023/12/optimizing-generative-ai-for-edge-devices)
- [Deploying ML Models on The Edge - @Microsoft](https://conferences.oreilly.com/artificial-intelligence/ai-eu-2019/cdn.oreillystatic.com/en/assets/1/event/299/Deploying%20machine%20learning%20models%20on%20the%20edge%20Presentation.pdf)
- [Fine-Tuning Small Language Models - Kili](https://kili-technology.com/large-language-models-llms/a-guide-to-using-small-language-models#fine-tuning-small-language-models)
- [AI on the edge: latest insights and trends @Qualcomm](https://www.qualcomm.com/news/onq/2023/09/ai-on-the-edge-the-latest-on-device-ai-insights-and-trends)
- [Small is the new big: the rise of small language models - Capgemini](https://www.capgemini.com/be-en/insights/expert-perspectives/small-is-the-new-big-the-rise-of-small-language-models/)
- [The 5 leading small language models of 2024: Phi 3, Llama 3, and more - DSDojo](https://datasciencedojo.com/blog/small-language-models-phi-3/)
- [7 Steps to Running a Small Language Model on a Local CPU](https://www.kdnuggets.com/7-steps-to-running-a-small-language-model-on-a-local-cpu)
### **Deep Dive Podcasts**
- [SmolLM2 Released: A Series (0.1B, 0.3B, and 1.7B) of Small Language Models for OnDevice Applications](https://www.youtube.com/watch?v=7J0PAffn-QU)
### **Books**
- [Machine Learning Systems - Vijay Janapa Reddi / Harvard (online & interactive book)](https://mlsysbook.ai/)
- [Edge-AI Books Collection - @cs-books](https://github.com/afondiel/cs-books/edge)
### Research Papers
- **[2024/08/30]** [Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone - Microsoft](https://arxiv.org/pdf/2404.14219)
- **[2024/06/16]** [Super Tiny Language Models](https://arxiv.org/pdf/2405.14159)
- **[2024/06/11]** [Small-E: Small Language Model with Linear Attention for Efficient Speech Synthesis ](https://arxiv.org/pdf/2406.04467)
- **[2023/04/28]** [EDGE IMPULSE: AN MLOPS PLATFORM FOR TINY MACHINE LEARNING](https://arxiv.org/pdf/2212.03332)