Flowise is an open-source, visual low-code platform for building LLM workflows and AI agents. It provides a drag-and-drop editor, a LangChain-compatible node library, and connectors for prompts, retrieval, tool calls, and multi-agent logic. One-sentence value: a vendor-neutral, extensible orchestration layer you can run under your control. It is aimed at

Argilla is an open-source, human-in-the-loop data curation and annotation platform for building, labeling, validating, and monitoring datasets used to train and maintain LLMs and NLP models. It provides a web UI, a Python SDK, a server API, and storage/connectors to assemble high-quality labeled datasets and feedback loops for training

MindsDB is an open-source AI data automation platform that lets teams build SQL-first AI queries, knowledge bases, and real-time answers over data spread across databases and applications. It exposes model training and inference via SQL and “AI tables,” reducing ETL and scaffolding for prototyping. It fits small, cost-aware teams, analytics

Label Studio is an open-source, multi-modal data labeling and annotation tool to create high-quality training data for machine learning and analytics. It provides configurable labeling interfaces, model-assisted pre-labeling, and export formats that turn human annotations into production-ready datasets. It is aimed at data scientists, ML and analytics engineers, and small

Ollama is a self-hosted platform to run large language models locally. It supports chat interactions with open models and can optionally use cloud-hosted models. It enables on-premises LLM inference to avoid cloud data egress, meet data residency and governance needs, and reduce reliance on external APIs. It targets individual developers

LiteLLM is an open-source LLM gateway and Python SDK that exposes a single OpenAI‑compatible API surface for completions, chat, embeddings and image generation while routing requests to many underlying providers. It provides an official proxy server (Docker images), a first‑class Python client, and features for routing, retries, budgeting,

Unstructured is an open-source plus commercial platform that ingests, partitions, and transforms heterogeneous documents (PDFs, images, HTML, Office files) into structured, AI-ready data. It focuses on semantic partitioning, modular ETL primitives, and built-in connectors so documents can be prepared for embeddings, retrieval, and other LLM-driven workflows. The platform targets engineering

AnythingLLM is an open-source, privacy-first platform for running and connecting large language models, whether locally or in the cloud. It combines document ingestion, vector search, RAG workflows, and a no-code agent builder so teams can stand up practical AI assistants without building everything from scratch. It’s aimed at developers,

Langfuse is an open‑source LLM observability platform for building, monitoring, debugging, and cost‑tracking large‑language‑model applications. It captures full traces of prompts, model inputs/outputs, and metadata so teams can reproduce behavior, evaluate changes, and control spending. It is aimed at engineering teams running LLMs in production

Open WebUI is a self-hosted web interface for running and managing large language models (LLMs) on your own machines or servers. It supports offline operation, progressive web app (PWA) behavior, and integrates with multiple model runners, including local backends and OpenAI‑compatible APIs. It is aimed at developers, researchers, and