Browse All AI Tools

Open-source visual RAG application builder

FlowiseAI provides a visual drag-and-drop builder specifically optimized for creating RAG applications. It supports multiple vector stores, embedding models, and LLM providers with an intuitive visual interface.

Build and share ML demos with a few lines of Python

Gradio enables ML practitioners to build interactive demos and web apps for their models with minimal Python code. It supports any model or API and provides shareable links for instant deployment.

High-performance text generation inference server

Text Generation Inference (TGI) by Hugging Face is a high-performance inference server for deploying LLMs. It provides continuous batching, tensor parallelism, and quantization for efficient model serving.

Pre-built LangChain application templates for common use cases

LangChain Templates provides production-ready starter templates for common AI application patterns including RAG chatbots, agents, and data processing pipelines. They accelerate development with best-practice implementations.

Open-source platform for ML lifecycle management

MLflow is the most popular open-source platform for managing the complete machine learning lifecycle. It provides experiment tracking, model registry, deployment, and LLM evaluation tools used by thousands of organizations.

World foundation models for physical AI and robotics

NVIDIA Cosmos provides world foundation models that understand physics, spatial relationships, and real-world dynamics. These models power simulation, synthetic data generation, and physical AI applications.

Content platform with AI-powered content operations

Sanity provides a content platform with AI capabilities for content generation, translation, and optimization. Its structured content lake enables AI to work with content meaningfully across channels and formats.

ML model registry and lifecycle management

W&B Model Registry provides comprehensive model versioning, lineage tracking, and lifecycle management for ML teams. It tracks every model from training through production with full reproducibility.