Dev Tools
Gradio
Build machine-learning web demos in Python within minutes — the fastest way to share a model with the world.
★ 43.7k starsApache-2.0
View guide →PostgresML — machine learning inside PostgreSQL with GPU acceleration: train, embed and run LLMs in SQL.
postgresml is an open-source dev tools project by postgresml (postgresml/postgresml on GitHub). It has 6,821 GitHub stars and 364 forks, is released under the MIT license, and was last updated on GitHub on 1 July 2025.
Decide cloud versus self-hosted, and pick a first workload: in-database training, embeddings/RAG, or LLM calls from SQL.
Clone postgresml/postgresml from GitHub (or use the install path in the run steps below) and skim its README for the parts relevant to your goal.
The fastest self-hosted path is the documented Docker image with a persistent volume; PostgresML Cloud skips installation entirely.
Connect with psql, run one in-database ML or embedding query on a sample table, and compare it with your current HTTP-serving approach.
Run machine learning and AI operations directly within PostgreSQL, keeping models and data together for privacy and simplicity.
Leverage GPU power for faster training computations and model inference inside the database.
Integrate and use state-of-the-art LLMs from Hugging Face in your SQL workflows.
Functions for chunking, embedding, ranking and transforming text, with vector search via pgvector integration.
A diverse set of classification and regression algorithms available in-database for classic ML tasks.
Positions inference as 8–40x faster than HTTP-based serving in its benchmarks, with horizontal scaling support.
Get the source from the official GitHub repository for postgresml.
git clone https://github.com/postgresml/postgresml.git
cd postgresmlStart the published image with its data volume and ports exactly as the README's self-hosted section shows.
docker run -it -v postgresml_data:/var/lib/postgresql -p 5433:5432 -p 8000:8000 ghcr.io/postgresml/postgresml:2.10.0 sudo -u postgresml psql -d postgresmlOnce connected, follow the README/docs examples to run an embedding, RAG or training query in SQL.
# Connect to the postgresml database and run a documented example query — see READMEThe PostgresML project, through the postgresml/postgresml repository, with the site and cloud at postgresml.org.
Yes — the project is open source under the MIT license as recorded for this listing. You can use, modify and distribute it under the MIT terms.
Self-hosting is free under MIT with no API key; PostgresML Cloud is the paid managed option. GPU features need GPU hardware you provide.
A separate API means network hops, duplicated data pipelines and another service to secure. PostgresML moves training and inference into Postgres itself — simpler data flow and lower latency in its benchmarks, in exchange for running ML inside your database tier.
Clone the repository (or use the documented installer), follow the install and configure steps on this page, and run the smallest real task first — connect with psql, run one in-database ML or embedding query on a sample table, and compare it with your current HTTP-serving approach.
Build machine-learning web demos in Python within minutes — the fastest way to share a model with the world.
From ticket to reviewed pull request — an autonomous coding pipeline on your machine.
GenBI for AI agents - governed text-to-SQL and dashboards through an open context layer
A GPU-accelerated, differentiable physics engine on NVIDIA Warp for robotics simulation.
The agent harness for complex codebases inside Codex — project memory, planning and verified completion.
grepai: semantic code search and call graphs for AI agents, running 100% locally on your machine.
The most-starred other Dev Tools projects in the LibreAI catalogue — the realistic shortlist if postgresml is not the right fit.
The open source coding agent
The model-definition framework behind modern AI — thousands of pretrained transformers for text, vision, and audio.
Anthropic's agentic coding tool that lives in your terminal and works across your whole codebase
Cross-platform desktop all-in-one manager for Claude Code, Codex, OpenCode and other coding agents
OpenAI's lightweight coding agent that runs locally in your terminal
Free and open — grab the source, star the repo, and start creating.