#Distillation

3 articles with this tag

Nvidia's Ziv Ilan on Faster Diffusion Models
AI Research

Nvidia's Ziv Ilan on Faster Diffusion Models

Nvidia's Ziv Ilan explains how to reduce diffusion model latency using quantization, caching, and distillation, plus the new FastGen library.

5 months ago
Ben Kunkle on Building Zed's Zeta2 Prediction Model
Artificial Intelligence

Ben Kunkle on Building Zed's Zeta2 Prediction Model

Ben Kunkle from Zed Industries explains the architecture and data pipeline for building Zeta2, an AI model that predicts code edits.

5 months ago
Cross-Architecture dLLM Distillation
AI Research

Cross-Architecture dLLM Distillation

TIDE framework enables cross-architecture distillation for diffusion large language models, achieving significant performance gains with smaller student models.

6 months ago