250 Open Source Apps
Ranked & Scored

A single-binary local LLM runner with a clean REST API for pulling, running, and serving GGUF models on CPU or GPU. Its standout trait is exceptional simplicity — one binary, one API, no dependency sprawl.

A unified LLM gateway/proxy that exposes an OpenAI-compatible API in front of 100+ model providers, with virtual keys, budgets, and routing logic. Standout trait is its genuinely stateless proxy design, making it one of the more horizontally-scalable apps in this batch.

A flexible data labeling tool supporting text, image, audio, and video annotation for ML training data, with configurable labeling interfaces. Standout trait is its template-driven UI that adapts to nearly any annotation task without custom code.

A ChatGPT-style web UI for local and remote LLM backends (Ollama, OpenAI-compatible APIs). Standout trait is fully local operation with no forced external API dependency.

A Python framework for packaging, serving, and scaling ML models as production APIs, with first-class support for building deployable containers ('Bentos'). Its standout trait is treating containerization as a core build artifact rather than an afterthought.

Open-source framework for building production AI agents with memory, plugins, and a built-in admin UI — the batteries-included alternative to wiring LangChain together yourself.

A full LLM application development platform with visual workflow building, RAG pipelines, and agent orchestration. Standout trait is enterprise-oriented breadth, delivered as a genuinely heavy multi-service stack.

Visual chatbot/conversational-form builder for building interactive flows without code, comparable to Typeform with logic branching. Multi-service Next.js architecture with a builder UI and a separate runtime viewer.

An MLOps framework for building portable, reproducible machine-learning pipelines that can run on your laptop or be pushed to cloud/Kubernetes orchestrators unchanged. Its standout trait is a pluggable 'stack' abstraction decoupling pipeline code from infrastructure.

Microsoft's toolset for building, testing, and orchestrating LLM-based application workflows (prompt chains, evaluation, tracing). Standout trait is strong tooling for evaluating and iterating on prompt flows, not just running them.

An open-source AI app builder — describe what you want, and it generates a working app. Notably, this instance runs on Nexlayer itself as the LLM backend.

An open-source platform for tracking ML experiments, packaging models, and managing a model registry across the ML lifecycle. Standout trait is broad framework-agnostic adoption as a de facto standard for experiment tracking.

A platform for capturing LLM production data and fine-tuning smaller, cheaper models to replace expensive frontier-model calls. Its standout trait is treating fine-tuning as a continuous data pipeline rather than a one-off training job.

An all-in-one RAG chat application that lets teams point an LLM at their own documents. Standout trait is bundling vector storage, document ingestion, and chat UI into one deployable unit.

A self-hostable AI assistant that indexes and searches personal notes, PDFs, and chat history. Standout trait is a strong semantic search layer over your own knowledge base rather than a generic chatbot.

Build, deploy, and run autonomous AI agents — one of the original projects that popularized agentic AI workflows, now a full platform with a marketplace and visual builder.

Open-source LLM chat interface supporting multiple model providers (OpenAI, Anthropic, and more) and plugins, built on Next.js with a polished, ChatGPT-like UX.

A drag-and-drop visual builder for LangChain-style LLM workflows. Standout trait is rapid prototyping of agent/RAG pipelines, at the cost of production-grade data and security hardening.

PrivateGPT lets you run document Q&A and RAG entirely offline against local LLMs, keeping data off third-party APIs. Its standout trait is a privacy-first design built around fully local inference.

An open-source RAG chatbot that lets users query their own documentation or file uploads through an LLM-backed Q&A interface. Standout trait is the batteries-included ingestion pipeline for turning docs into a searchable vector index.

A self-hosted, OpenAI-API-compatible inference server for running open-weight LLMs, image, and audio models locally. Standout trait is broad model-format support (GGUF, ONNX, diffusers) behind one consistent API surface.

A React + Node framework/SDK for embedding AI copilots (chat, autocomplete, in-app agents) into existing web applications. Its standout trait is tight React hook integration that lets app state be read/written by an LLM agent with minimal glue code.

A locally-run frontend for LLM-based character roleplay and chat, supporting many backend model providers. Its standout trait — and biggest weakness — is that it ships with no authentication by default, designed purely for single-user local use.

A Gradio-based local chat UI for running LLMs (llama.cpp, ExLlama, Transformers, GGUF) on your own GPU/CPU. Its standout trait is backend flexibility — swap inference engines and models without changing the UI.

A node-graph based UI for building and running Stable Diffusion and other image/video generation pipelines visually. Standout trait is its highly composable workflow-as-graph model, which power users export and share as JSON.

A proxy layer that reverse-engineers free access to various LLM providers' web interfaces. Standout trait is that it is explicitly built around evading rate limits and ToS rather than being a legitimate self-hostable service.