Machine LearningDev30 min reading time

How to Run Underdog Saluki 27B: 2-Bit Qwen3.8-27B Tool-Calling Guide

MarkTechPost
Read the full article
Underdog Saluki 27B is a 7.9 GB quantized version of Qwen3.8-27B that maintains tool-calling capabilities at 2-bit precision and outperforms the full 54 GB model on a 120-task benchmark. The tutorial compares Saluki to ISTA-DASLab's GSQ-RCO quantization using llama.cpp, evaluating memory, speed, and task performance.

More in Machine Learning

Machine Learning6 min read

Micro1 is making a $1 billion bet on AI's latest data craze

Business Insider
Machine Learning2 min read

Samsung aims to cut chip material development time using AI

Yonhap News Agency
Machine Learning5 min read

PyTorch Conference 2026: Google and Meta Open-Source TorchTPU for Native PyTorch on TPUs

InfoQ (AI, ML & Data)