Zive

American Dream Jobs is an open-source platform designed to help users discover high-paying, AI-resilient careers. It provides personalized job recommendations, career path comparisons, and local job postings. The platform also offers information on training programs and AI resilience scores to understand career trajectories.

Roboflow is a computer vision platform that provides tools for building and deploying computer vision models. It allows users to manage image data, annotate datasets, and train models with one-click training. Over 1,000,000 developers use Roboflow to deploy models via API or to the edge.

MultiLLM is an AI model comparison tool that allows users to query multiple LLMs simultaneously. It supports models like GPT, Claude, and Gemini, providing side-by-side comparisons of their responses. Users can pin, search, and revisit conversations. It is designed for writers, programmers, and researchers.

Seekr delivers trusted AI solutions for mission-critical environments. It helps government and critical infrastructure leaders make accurate decisions. SeekrFlow is an end-to-end platform with content-scoring capabilities. The toolset allows users to build domain-specific large language models and AI agents. Seekr addresses bias, data integrity, and model complexity.

awesome-llms-fine-tuning is a curated collection of resources for fine-tuning Large Language Models (LLMs). It includes tutorials, papers, tools, and best practices. This repository is designed for ML practitioners and researchers. It provides resources to adapt pre-trained models to specific tasks.

Trinity-RFT is a framework designed for reinforcement fine-tuning (RFT) of large language models (LLM). It is a general-purpose tool intended to be flexible and scalable. Trinity-RFT allows developers to fine-tune LLMs using reinforcement learning techniques.

smolGPT is a minimal PyTorch implementation for training small language models from scratch. It is designed for educational purposes and simplicity, featuring efficient training and modern sampling techniques. The tool includes Flash Attention and RMSNorm for optimized performance. It is available on GitHub.

robustlearn is a unified library for research on robust machine learning. It supports adversarial/backdoor attack and defense, and out-of-distribution generalization. The library facilitates research in responsible AI and the development of robust machine learning models. It is designed to be extensible and supports various machine learning tasks.

spaCy is a Python library for advanced Natural Language Processing (NLP). It is built on research and designed for use in real products. spaCy comes with pre-trained pipelines and supports tokenization and training for over 70 languages, featuring neural network models.