# Edge device deployment

PP-OCRv3_mobile_rec is a lightweight text line recognition model developed by the PaddleOCR team. It uses the SVTR algorithm and supports Chinese and English recognition, especially focusing on Chinese scenarios.

The Holo1-7B GGUF model is part of the Surfer-H system and is suitable for multimodal tasks such as visual document retrieval. It is particularly good at web page interaction and network monitoring, and can achieve high accuracy at a low cost.

Llama-3.1-Nemotron-Nano-VL-8B-V1 is an advanced document intelligent vision-language model that can query and summarize images and videos, and supports multi-environment deployment.

Llama-3.1-Nemotron-Nano-4B-v1.1 is a large language model optimized based on Llama 3.1, achieving a good balance between accuracy and efficiency. It is suitable for various scenarios such as AI agents and chatbots.

AceMath-RL-Nemotron-7B is a mathematical reasoning model trained entirely through reinforcement learning. It is trained based on Deepseek-R1-Distilled-Qwen-7B and performs excellently in mathematical reasoning tasks. It also has certain generalization ability in coding tasks.

D-FINE is a powerful real-time object detector that achieves exceptional localization accuracy by redefining the bounding box regression task in DETR models.

D-FINE is a real-time object detection model that achieves exceptional localization accuracy by redefining the bounding box regression task.

Qwen2.5-VL-3B-Instruct is a 3B-parameter multimodal model supporting image-text generation tasks, specifically optimized for vision capabilities in llama.cpp.

GGUF quantized version of Gemma 3 with 27B parameters, supporting image-text interaction tasks

RT-DETRv2 is an improved real-time object detection model based on the DETR architecture, optimizing detection performance through innovations like selective multi-scale feature extraction and dynamic data augmentation.

RT-DETRv2 is an improved version of the real-time object detection Transformer model, enhancing performance through multi-scale feature extraction and optimized training strategies.

Qwen2-Audio is an advanced small-scale multimodal model that supports audio and text input, enabling voice interaction without relying on speech recognition modules.