All Open Source Apps

An Apple-Silicon-native LoRA fine-tuning tool for Gemma on text, image, and audio data — with a wizard CLI, live browser-based training visualizer, and streaming from GCS/BigQuery for datasets too large for local disk.

Rasa Open Source is a Python machine learning framework for building contextual, multi-turn chatbots and voice assistants that understand natural language and maintain conversation state.

Real-time NVIDIA GPU monitoring dashboard — self-hosted, scale from 1 to 100+ GPUs with sub-second metrics and no cloud dependency.

Morphik is an AI-native ingestion and retrieval engine that lets developers store, search, and reason over visually rich documents — scanned PDFs, manuals, slides, and video — without duct-taping together OCR, an embedding model, and a vector database.

A fully local web research assistant that iteratively searches, summarizes, and refines markdown reports using any Ollama or LMStudio model—no cloud or API keys required.

Build autonomous and human-in-the-loop multi-agent AI systems with a layered, event-driven Python and .NET framework pioneered at Microsoft Research.

TensorZero unifies the LLM gateway, observability, evaluation, optimization, and experimentation stack behind a single OpenAI-compatible API, built in Rust for sub-millisecond p99 latency.

Run ML training and LLM inference natively inside PostgreSQL with GPU acceleration — no data movement required.

One dashboard to monitor, trigger, and cancel CI/CD pipelines across GitHub, GitLab, Jenkins, Tekton, Buildkite, ArgoCD, and Bitbucket — desktop, Docker, or iOS.

An open-source build system that turns written specs and architecture into working code — an LLM generates code under tight constraints, task by task with narrow context windows, instead of attempting an entire codebase at once.

Self-hosted, open-source Vercel alternative that deploys Python, Node.js, PHP, and any Docker-compatible app from a Git push, with zero-downtime rollouts and real-time logs.

Baidu's open-source OCR model for "one-shot, long-horizon parsing" — published with an arXiv paper, a Hugging Face model release, and vLLM inference support, aimed at document understanding beyond typical page-by-page OCR limits.

All-in-one self-hostable platform for hybrid search, RAG, recommendations, and analytics built on Rust and Qdrant.

Give an AI agent a real LLM training setup and let it experiment autonomously overnight — you wake up to a log of experiments and (hopefully) a better model.

Self-hosted backup automation for Ubuntu and Debian servers with S3 storage, SSH-based file and database backup, and multi-channel notifications.

A local-first AI memory sidecar that extracts entities, builds a persistent knowledge graph, and injects ranked context into any LLM pipeline — no cloud required.