Back to AIBriefs
AnalysisAI Models

Qwen3.6-27B quantization benchmark compares quants from Q8 to Q2

Qwen3.6-27B quantization benchmark compares quants from Q8 to Q2View on Reddit

Reddit user bobaburger benchmarks Qwen3.6-27B quantizations (Q8 to Q2) using llama.cpp perplexity. Compares quants from unsloth, mradermacher, IQ4_XS by cHunter789 and Ununnilium.

·
15 days ago
Qwen3.6-27B quantization benchmark compares quants from Q8 to Q2 — AIBriefs