Saluki 27B Local Test | Same As Qwen3.8 27B with Less Memory? | 16GB Local AI Setup

What if you can run 27B model that fits in ~11GB and is almost as smart as the original? Is it too good to be true? Let's test Saluki 27B (heavily quantized version of Qwen3.8 27B) and find out! DaBench run: https://dabench.ai/model/underdog-saluki-27b-1-0 Announcement: https://underdog.ai/saluki Weights: https://huggingface.co/ConwayResearch/Underdog-Saluki-27B-1.0 AI Academy: https://MLExpert.io Work with me: https://mlexpert.io/consulting DaBench (open weight LLM benchmarks): https://dabench.ai LinkedIn: https://www.linkedin.com/in/venelin-valkov Follow me on X: https://twitter.com/veneli