<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>NVIDIA Nemotron on TechLLM</title><link>https://techllm.github.io/tags/nvidia-nemotron/</link><description>Recent content in NVIDIA Nemotron on TechLLM</description><generator>Hugo</generator><language>ko</language><lastBuildDate>Sat, 23 May 2026 18:09:00 +0900</lastBuildDate><atom:link href="https://techllm.github.io/tags/nvidia-nemotron/index.xml" rel="self" type="application/rss+xml"/><item><title>Nemotron-Labs Diffusion: LLM 생성 속도를 바꾸는 세 가지 모드</title><link>https://techllm.github.io/posts/nemotron-labs-diffusion/</link><pubDate>Sat, 23 May 2026 18:09:00 +0900</pubDate><guid>https://techllm.github.io/posts/nemotron-labs-diffusion/</guid><description>NVIDIA의 Nemotron-Labs Diffusion은 기존 LLM의 순차 생성 한계를 줄이기 위해 자동회귀, 디퓨전, self-speculation을 한 모델 안에 묶은 새로운 언어 모델 계열이다.</description></item></channel></rss>