Qwen3.6 27B

on 2× NVIDIA RTX 3090 Ti · llama.cpp · 196,608 ctx

Jun 23, 2026

Throughput

100.0 t/s gen

Quant

Q8_0 (GGUF)

Summary

User reports tensor split-mode raising throughput from 70+ t/s to 100+ t/s, with a peak of 130 t/s. Power draw is 750W+.