physicsnemo
Open-source deep-learning framework for building, training, and fine-tuning deep learning models using state-of-the-art Physics-ML methods
- 3.5k
- Python
- Apache-2.0
- 4 h ago
Open-source deep-learning framework for building, training, and fine-tuning deep learning models using state-of-the-art Physics-ML methods
Self-hosted previews, intro/credits detection and cross-server reuse for Plex, Emby and Jellyfin, with GPU acceleration and remote processing. Includes Plex chapter thumbnails and CPU loudness analysis; Sonarr/Radarr automation.
Securade.ai HUB - A generative AI based edge platform for computer vision that connects to existing CCTV cameras and makes them smart.
An end-to-end agent project for GPU kernel implementation, analysis, profiling, and iterative optimization. It helps an agent turn PyTorch logic or an existing kernel into a high-performance GPU kernel through a structured, profile-driven workflow.
Production-ready C++/TensorRT inference engine for RF-DETR. Object detection and instance segmentation with FP32/FP16/INT8 support. Optimized for NVIDIA GPUs, Jetson (Orin, AGX Thor).
40x faster AI inference: ONNX to TensorRT optimization with FP16/INT8 quantization, multi-GPU support, and deployment
Pre-driver FMLA32 Tensor Core unlocker and PTX benchmark console for GeForce RTX 40-series, RTX 30-series and RTX 20-series on Windows 10/11 (UEFI USB) and Linux (gpu_unlock.ko), with friendly UI for Win/Linux. Restores full 2x FP16/BF16/TF32->FP32 accumulate rate with automatic hardware profiling, pre-flight ROM checks, and zero VBIOS flashing.
SORDI dataset has per frame annotation file in json format. Following tools create a COCO style annotation out of it. Thus the SORDI data can be easily fed into COCO style training pipelines.
A comprehensive multimodal vision-language model application supporting both image and video inference tasks. This application integrates four specialized models for document processing, OCR, spatial reasoning, and video understanding.
A Gradio-based demo for end-to-end vision-to-speech inference: Extract text or descriptions from images using Qwen2.5-VL-7B-Instruct, then convert to natural speech audio via Microsoft VibeVoice-Realtime-0.5B.
V100-optimized mixed precision for Qwen Image in ComfyUI — 1.43x faster sampling on NVIDIA Tesla V100 32GB (Volta/sm_70).
batch-encode your ripped DVDs and Blurays really fast with nvenc_av1!