EU AI Act OpenRAG: 933 legally structured chunks and BGE-M3 embeddings in one SQLite file [P]
Summary
A resource containing 933 legally structured chunks of the EU AI Act along with BGE-M3 embeddings, packaged in a single SQLite file for easy use in retrieval-augmented generation systems.
Similar Articles
HKUDS/RAG-Anything
HKUDS released RAG-Anything, an open-source all-in-one multimodal retrieval-augmented generation framework based on LightRAG.
LiquidAI/LFM2.5-Embedding-350M
Liquid AI releases LFM2.5-Embedding-350M, a dense bi-encoder for multilingual retrieval supporting 11 languages, as a drop-in replacement for RAG pipelines.
Models for minimalist RAG: B1ade 335M Embedding and 1B Parameter Small Language Models
Presents B1ade, a minimalist RAG architecture with a 335M zero-training embedding model and a 1B SLM trained via GRPO on 723M tokens, showing emergent attribution behavior and competitive QA performance without large-scale pretraining.
Ant Group put a 124B model under plain MIT, not one of those "community" licences
Ant Group's inclusionAI released Ling-3.0-flash, a 124B-parameter (5.1B active) model under a plain MIT license with no revenue caps or usage restrictions, and it's available on OpenRouter with a free tier.
I built a new type of AI tool; it generates 3D objects composed of their constituent parts (instead of the monolithic solid blobs all 3D AI generators produce).
A new AI tool generates 3D objects by generating code, resulting in objects with separate, functional parts rather than monolithic blobs. It is free and open-source on GitHub.