Julia Turc

The physics behind Flow Matching models

Julia Turc

Subham Sahoo The Future of Discrete Diffusion

Julia Turc

How is hardware reshaping LLM design

Julia Turc

How did diffusion LLMs get so fast

Julia Turc

An image is worth NxN words Diffusion Transformers (ViT, DiT, MMDiT)

Julia Turc

Why are Transformers replacing CNNs

Julia Turc

I asked them to show me their RAG pipeline...

Julia Turc

David Petrou Building social AI after 17 years at Google

Julia Turc

Transformers & Diffusion LLMs What's the connection

Julia Turc

Text diffusion A new paradigm for LLMs

Julia Turc

Hierarchical Reasoning Model Substance or Hype

Julia Turc

The physics behind diffusion models

Julia Turc

Reverse-engineering GGUF Post-Training Quantization

Julia Turc

Training models with only 4 bits Fully-Quantized Training

Julia Turc

The myth of 1-bit LLMs Quantization-Aware Training

Julia Turc

How LLMs survive in low precision Quantization Fundamentals

Julia Turc

Knowledge Distillation How LLMs train each other

Julia Turc

Mixture of Experts How LLMs get bigger without getting slower

Julia Turc

Llama 4 Explained Architecture, Long Context, and Native Multimodality

Julia Turc

DeepSeek's GRPO (Group Relative Policy Optimization) Reinforcement Learning for LLMs

Julia Turc

Proximal Policy Optimization (PPO) for LLMs Explained Intuitively

Julia Turc

Tülu 3 from AI2 Full open-source fine-tuning recipe for LLMs

Julia Turc

8 Timeless tips for training LLMs Become a better ML engineer

Julia Turc

How does DeepSeek actually work Full technical review

Julia Turc

Discover How LLMs Work by Dissecting Llama

Julia Turc