📚 Introduction to Modern Statistics - A college-level open-source textbook with a modern approach highlighting multivariable relationships and simulation-based inference.

Python package for the creation, manipulation, and learning of Causal DAGs

Full in-browser Semantic Search with Huggingface Transformers.js and ElectricSQL's PGlite!

Create a smart video application using the Intel Distribution of OpenVINO toolkit. The toolkit uses models and inference to run single-class object detection.

TensorFlow as a Service, a general purpose framework to serve TF models.

A Tool to Build NodeJS and Web Based Expert Systems

The client for the Symmetry peer-to-peer inference network. Enabling users to connect with each other, share computational resources, and collect valuable machine learning data.

AI in A Box

A web-based memory usage and performance calculator for Huggingface GGUF models

A GPU defined in software. Runs Llama 3.2 1B at 3.6 tok/sec. Zero dependencies.

Provider-pluggable orchestration runtime for multi-model AI inference. ( Sakana Fugu style )

🔀 Bedrock Proxy Endpoint ⇢ Spin up your own custom OpenAI API server endpoint for easy AWS Bedrock inference (using standard baseUrl, and apiKey params)

AIME API Server - Scalable AI Model Inference API Server

Scalable AI Inference Server for CPU and GPU with Node.js | Utilizes llama.cpp and parts of llamafile C/C++ core under the hood.

Object detection inference with Roboflow Train models on NVIDIA Jetson devices.

Util for casting any object to specified model with typescript type inference

Personal Reasoning Engine — Gemma 4 powered Agentic assistant running locally and optimized for Apple Silicon with Windows compatiability. Built in objective c, rich CLI, rich GUI, zero cloud dependency.

Bayesian learning and inference with WebPPL.

🪨 Bedrock Wrapper is an npm package that simplifies the integration of existing OpenAI-compatible API objects with AWS Bedrock's serverless inference LLMs.

Free OpenAI-compatible LLM endpoint. Zero API keys, zero signup, zero user compute. Aggregates keyless public endpoints (Pollinations, ApiAirforce) with auto-failover, client-side queueing, and ad/notice filtering.