MCP server for the DeepGHS anime AI ecosystem. Dataset discovery, tag search, character dataset finder, and training pipeline code generation.

Code for the paper comparing LoRA on Vision Transformer vs. traditional fine-tuning for CNNs on MNIST, Fashion-MNIST, and CIFAR-10.

LoRA-based fine-tuning and RAG to build on-premises chatbots using small open-weight language models

A tool to analyse images and captions for LoRa datasets

Real-time Moroccan Darija Speech-to-Text transcriber and LLM assistant. Built with an Angular frontend, FastAPI Python backend, and a fine-tuned Qwen3-ASR model via streaming WebSockets. Fully containerized with Docker.

Train a personal FLUX.1-dev LoRA on your own photos and generate professional headshots locally. 8 GB VRAM friendly.

musubi-tuner GUI (Krea2 only)

Automates the setup and installation of invoke-training on Windows, Mac, and Linux.

Local AI workstation dashboard — GPU monitoring, training jobs, ComfyUI/SwarmUI control, model discovery hub

Lofi's Lora Data Prep tool with a GUI to make curating your images easy.

LoRA素材作成向けの画像トリミングツール。ZIP/RAR/7z内の画像確認、固定サイズでの切り取りや背景除去に対応。

Elixir SDK for the Tinker ML platform—LoRA training, sampling, and service orchestration built on OTP, Finch, and telemetry.

Complete guide + scripts for training HunyuanVideo 1.5 I2V LoRAs on AMD GPUs (ROCm). Battle-tested on RX 9700 XT.

Local LLM + ComfyUI + Open WebUI on a single NVIDIA GPU. llama.cpp proxy with auto GPU mode swapping (LLM / image-gen / LoRA training). Benchmarking + model-preset switching.

An Arabic News Credibility Checking platform that involved finetuning an LLM, building an API, and hosting it in the cloud using Azure

12 production-tested ComfyUI workflows — AI UGC videos, trend copying, character generation, music, voice cloning, and more

AI-Powered Video Processing Tool for LoRA Training Dataset Creation.

BoneMemory: Universal Async Core for AI-Toolkit. The hardware-agnostic VRAM manager for every model: Image, Video, and Audio. Whether you’re training a Rank 16 LoRA on an entry-level GPU or pushing Rank 1024 on an RTX 4090, BoneMemory eliminates memory bottlenecks. Architecture that makes any hardware punch above its weight. Zero OOM.

An experiment demonstrating instruction-following distillation, enabling the transfer of knowledge from large language models (Teachers) into smaller, efficient models (Students) using LoRA-based fine-tuning and structured LLM-based evaluation.