Stijn

agentic-cursorrules. A practical approach to managing multiple AI agents in Cursor through strict file-tree partitioning and domain boundaries.

ui-screenshot-to-prompt. ui-screenshot-to-prompt is an AI-powered tool that analyzes UI images to generate detailed prompts for AI coders. It uses computer vision and natural language processing to break down UI components, analyze design patterns, and create comprehensive descriptions for reproducing the design. Very useful for Bolt.new, v0 and other upcoming SaaS

modernbert-finetune. Fine-tune ModernBERT with custom tokenizers, curriculum learning, and next-gen optimizers.

grpo-optuna. Optimizing Causal LMs through GRPO with weighted reward functions and automated hyperparameter tuning using Optuna

automatic-learning-amplifier. MLX-based QA pair generator and LLM finetuning tool in Streamlit

fastapi-doctor. Rust-native dev-tool for Agents that scores FastAPI applications, diagnoses API issues, and helps agents understand, validate, and improve it's endpoints faster. Audit 250k loc within half a second.

docs-to-eval. docs-to-eval builds tailor-made evaluation sets from raw documentation and scores model outputs with the right verification strategy. It combines lightweight FastAPI services, reusable Python components, and optional agentic workflows.

RAG-finetuned-embeddings. A modular RAG system implementing Hybrid Retrieval (RRF), Semantic Chunking, and Contrastive Fine-tuning.

EasyDeL-GSPO. Accelerate your training with this open-source library. Optimize performance with streamlined training and serving options with JAX. 🚀

sentence-improver. AI-powered sentence improvement web app using Anthropic API

firecrawl. 🔥 Turn entire websites into LLM-ready markdown or structured data. Scrape, crawl and extract with a single API.

finetune-zamba2. finetune-zamba2

code-fusion. Select SoTA models and let a LLM decide what's working best

lora-selector. LoRA selector using fine-tuned embeddings and TinyLlama for specialized inference

agno. High-performance runtime for multi-agent systems. Build, run and manage secure multi-agent systems in your cloud.

open-webui. User-friendly AI Interface (Supports Ollama, OpenAI API, ...)

mergekit. Tools for merging pretrained large language models.

UIED. An accurate GUI element detection approach based on old-fashioned CV algorithms [Upgraded on 5/July/2021]

EasyLM. Large language models (LLMs) made easy, EasyLM is a one stop solution for pre-training, finetuning, evaluating and serving LLMs in JAX/Flax.

s-smits.

Zamba2. PyTorch implementation of models from the Zamba2 series.

multimodal-disaster-response. Embeddings-based disaster response predictor with Gradio interface for text analysis.

gitignore. A collection of useful .gitignore templates

awesome-o1-prompts. Curated list of chatgpt prompts from the top-rated GPTs in the GPTs Store. Prompt Engineering, prompt attack & prompt protect. Advanced Prompt Engineering papers.

ScandEval. Evaluation of language models on mono- or multilingual tasks.

fabric. fabric is an open-source framework for augmenting humans using AI. It provides a modular framework for solving specific problems using a crowdsourced set of AI prompts that can be used anywhere.

mlx-vlm. MLX-VLM is a package for running Vision LLMs locally on your Mac using MLX.

axolotl. Go ahead and axolotl questions

transformers. 🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.