Running Qwen 3.8–27B locally with Unsloth and DeepSeek Harness on an RTX 3090 (24 GB) under Windows 11.
Par Jacques Gariépy • Guide technique, retour d'expérience, dépannage Windows pas-à-pas et utilisation Web & CLI. Table des Matières Introduction & Architecture Globale Pourquoi ce Setup ? (RTX 3090 24 Go + UD-Q4_K_XL) Comment Obtenir & Générer vos Clés d'Accès Dépannage & Installation d'Unsloth Studio : Le Bug SSLKEYLOGFILE Installation & Compilation de DeepSeek Harness Démarrage du Serveur…
A technical guide provides a setup for running a local, autonomous software engineering agent on a machine with a NVIDIA RTX 3090 graphics card. The setup combines DeepSeek Harness, Unsloth Engine, and Qwen 3.8-27B model to achieve high performance. The guide explains how to obtain access keys, install and configure the software, and troubleshoot common issues, including an SSLKEYLOGFILE bug on Windows. The setup uses 23.5 GB of the RTX 3090's 24 GB VRAM, leaving no need for swapping to slower RAM.
Written by urgent.news from Dev.to's report — not a translation of it. Machine-written — may contain errors; check the original before relying on it.
