# Lightweight Model

LiteAvatar is an audio-driven real-time 2D avatar generation model primarily designed for real-time chat scenarios. Through efficient speech recognition and viseme parameter prediction technology combined with a lightweight 2D face generation model, it achieves 30fps real-time inference using only CPU. Key advantages include efficient audio feature extraction, a lightweight model design, and mobile device-friendly support. This technology is suitable for real-time interactive virtual avatar generation scenarios such as online meetings and virtual live streaming. It was developed based on the need for real-time interaction and low hardware requirements. Currently, it is open-source and free, positioned as an efficient, low-resource-consuming real-time avatar generation solution.

SmolLM2 is a series of lightweight language models, featuring versions with 135M, 360M, and 1.7B parameters. These models effectively handle a wide range of tasks while maintaining a lightweight profile, particularly for device deployment. The 1.7B version shows significant improvements over its predecessor, SmolLM1-1.7B, in instruction-following, knowledge, reasoning, and mathematics. It has been trained on multiple datasets, including FineWeb-Edu, DCLM, and The Stack, and has undergone Direct Preference Optimization (DPO) using UltraFeedback. The model also supports tasks such as text rewriting, summarization, and functional invocation.

Gemma-2-9b-it is a series of lightweight, state-of-the-art open models developed by Google, built upon the same research and technology as the Gemini model. These models are text-to-text decoder-only large language models, offered in English and suitable for a variety of text generation tasks, including question answering, summarization, and reasoning. Due to their relatively smaller size, they can be deployed in resource-limited environments such as laptops, desktops, or personal cloud infrastructure, making advanced AI models more accessible and fostering innovation.

Gemma 2 is a series of lightweight, advanced open models developed by Google, built upon the same research and technology as the Gemini model. They are text-to-text decoder-only large language models, offering an English-only version with open weights, suitable for both pre-training variations and instruction-tuning variations. The Gemma model is well-suited for a wide range of text generation tasks, including question answering, summarization, and reasoning. Its relatively small size enables deployment in resource-limited environments like laptops, desktops, or your own cloud infrastructure, democratizing access to advanced AI models and fostering innovation for everyone.

MotionFollower is a lightweight score-guided diffusion model designed for video motion editing. It utilizes two lightweight signal controllers to separately control pose and appearance, avoiding heavy attention computations. The model employs a dual-branch architectural design based on the score-guided principle, including reconstruction and editing branches, significantly enhancing its ability to model texture details and complex backgrounds. Experiments demonstrate that MotionFollower reduces GPU memory usage by approximately 80% compared to the state-of-the-art motion editing model MotionEditor, while providing superior motion editing performance. It also exclusively supports extensive camera movement and complex actions.