python

Research experiments exploring uncommon quant techniques.

mindspore implementation of transformers

Example ChatGPT chatbots using Langchain and OpenAI

Synthetic test data that hits the numbers you declare, exactly. Multi-table with verified foreign-key integrity, deterministic, no model in the data path. Python + MCP server.

A chat-based recommendation application that revolutionizes the culinary experience.

Web app enabling users to either record or upload audio files. Then utilizing OpenAI API (Whisper, GPT4) generates transcriptions, summaries, fact checks, sentiment analysis, and text metrics. Users can also intelligently chat about their transcriptions with a GPT4 chatbot. Data is stored relationally in SQLite and also vectorized in Pinecone.

Local-first, open-source YouTube dubbing: clones the original voice into another language, time-synced. Chatterbox voice cloning + faster-whisper + NLLB, with multi-voice, burned subtitles and optional Wav2Lip lip-sync. Runs free on your own machine via a simple CLI.

Local-first media asset manager with AI semantic search — reads pro-camera metadata (Sony XAVC sidecar, FX-series) and 360 footage (Insta360/GoPro Max) that consumer DAMs drop. FFmpeg + ChromaDB + local LLM.

52-week journey from QA/SDET to GenAI Testing - learning in public with weekly mini-projects, code, and honest documentation of struggles and wins.

Unofficial Python API for Perchance.

Build Recommender System with PyTorch + Redis + Elasticsearch + Feast + Triton + Flask. Vector Recall, DeepFM Ranking and Web Application.

The most feature-complete local AI workstation. Multi-GPU inference, integrated Stable Diffusion + ADetailer, voice cloning, research-grade ELO testing, and tool-calling code editor. 100% local. Zero subscriptions. Your GPUs deserve better.

text to image generation with stable diffusion

CLI for audio, video, and text transcription with ASR providers and LLM-powered summarization via local or cloud backends.

Local AI chatbot for Matrix with infinite personalities, tool calling, and MCP — powered by Ollama

Just say what you want. In your terminal.

Too many docs? Quickly search over any PDF or Markdown documents

The repository explores various RAG techniques, including implementation guides, use cases, and best practices. Each article is designed to help researchers, developers, and enthusiasts understand and implement RAG systems efficiently.

The smallest possible LLM API

SPARC is an intelligent ROS2 Humble-based pick-and-place robotic system integrating MoveIt, YOLOv8-OBB, and Isaac Sim. It enables real-time object detection, motion planning for Franka Panda and UR5 arms, and natural language control via an Ollama-powered chatbot—making robotic manipulation smart, interactive, and human-friendly.