---
format: "aidr-story-markdown/v1"
id: "57571dc1d784b80ddfca60c97b8a3273d3fadd82358c3caa0615fd40f11a42d1"
canonical_url: "https://aidr.today/57571dc1?lang=en"
title: "How to Run NVIDIA Dory 4B with llama.cpp: Loop by Loop Guide"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-11-07T19:22:44.000Z"
category: "Frameworks"
topics: ["nvidia","llm","framework"]
source_urls: ["https://www.marktechpost.com/2026/11/07/coding-guide-nvidia-dory-4b-llama-cpp/"]
summary: "A guide is available for running NVIDIA's Dory 1.0 4B looped language model with llama.cpp. The guide covers building the pinned llama.cpp branch, analyzing loop-by-loop compute costs, testing memory caching, and running paired evaluations."
---

# How to Run NVIDIA Dory 4B with llama\.cpp: Loop by Loop Guide

> [Open the canonical story](<https://aidr.today/57571dc1?lang=en>)

**Published:** 2026-11-07T19:22:44.000Z
**Category:** Frameworks
**Topics:** nvidia, llm, framework

## Summary

A guide is available for running NVIDIA's Dory 1\.0 4B looped language model with llama\.cpp\. The guide covers building the pinned llama\.cpp branch, analyzing loop\-by\-loop compute costs, testing memory caching, and running paired evaluations\.

## Sources

- [Story source](<https://www.marktechpost.com/2026/11/07/coding-guide-nvidia-dory-4b-llama-cpp/>)