4 repositorios
Mounting host directories to store model weights and caches for persistent development environments.
Distinct from Model Weight Management: Focuses on directory mounting for cache persistence, distinct from general weight management.
Explore 4 awesome GitHub repositories matching artificial intelligence & ml · Cache Persistence Mounts. Refine with filters or upvote what's useful.
Detectron2 is a PyTorch computer vision framework and visual recognition platform designed for training and deploying models for object detection, image segmentation, and visual recognition. It provides a research-oriented environment for training complex vision models with multi-GPU acceleration. The project includes a specialized object detection library for identifying and locating multiple objects via bounding boxes, as well as an image segmentation toolkit for creating pixel-level masks through instance, semantic, and panoptic segmentation. Additionally, it features a human pose estimati
Implements host directory mounting to prevent repeated downloads of large pre-trained model files.
Sglang is a high-performance inference engine and serving system designed for large language and multimodal models. It provides a programmable interface for orchestrating complex generation workflows, enabling developers to coordinate multi-turn dialogues, tool invocations, and reasoning chains through a domain-specific language. The platform is built to support production-scale deployments, offering an OpenAI-compatible API that allows for integration with existing application ecosystems. The system distinguishes itself through a disaggregated architecture that separates compute-intensive pr
Mounts host directories to store model weights, preventing redundant downloads.
Earthly is a containerized build system and Docker build framework designed for creating reproducible build pipelines. It ensures environment consistency by executing every build step inside an isolated container, combining the isolation of container images with dependency tracking and parallel execution. The system differentiates itself through a focus on hermeticity and multiplatform support, allowing for the generation of container images targeting multiple CPU architectures within a single execution flow. It maintains a hermetic build environment by isolating network access and utilizing
Declares persistent cache mounts using a dedicated command to optimize incremental builds.
This project is a foundational boilerplate and starter framework designed for building content-heavy websites and blogs. It functions as a static site generator that transforms source files, such as Markdown and various template formats, into pre-rendered HTML during the build process. By eliminating the need for client-side processing, it produces high-performance, secure, and accessible web output ready for deployment to any standard hosting environment. The framework distinguishes itself through a robust build-time pipeline that manages complex content structures and asset optimization. It
Retains temporary data folders between build cycles to speed up subsequent builds on hosting providers.