Stories by omneity

Tokenization Is Killing Our Multilingual LLM Dream

I stopped trusting the official Wikipedia dataset, and what I did about it

A Wordle for the Worldle

Show HN: Wikilangs Games – Wordle-like for 300 Languages

Picomon 0.2.0: From AMD Crash Fix to GPU Monitoring That Doesn't Suck

Show HN: Picomon, a minimal TUI monitor for AMD GPUs

Show HN: BorgLLM, Zero-config LangChain Client, key rotation and rate limit mgmt

Show HN: BorgLLM, Zero-config LangChain Client, key rotation and rate limit mgmt

Meta Synthetic Data Kit

Aana SDK, a framework for building AI enabled multimodal applications

Getting Perfectly Structured Data from LLMs

Finding Moroccan Arabic (Darija) in the Fineweb 2 Dataset

Show HN: Datapluck, Portability Tool for HuggingFace Datasets

Show HN: Datapluck, Portability Tool for HuggingFace Datasets

Copied Act would make removing AI digital watermarks illegal

An African Superstate: The East African Federation (2022)

Mixtral 8x7B on 24GB VRAM (2-bit quantized)

OpenAI RAG: A Step Forward or a Step Too Far?

1 2