llamaperf

Qwen3.8 27B Vision

on 4× AMD RX 580 8GB + AMD RX 570 8GB + NVIDIA GTX 1660 Super 6GB + NVIDIA GTX 1070 8GB · Slipstream

Oct 23, 2026
Throughput
3.8 t/s gen

Use cases

vision

Summary

User reports Qwen3.8 27B Vision at 3.82 t/s on a four-GPU mixed-vendor node (RX 580 8GB, RX 570 8GB, GTX 1660 Super 6GB, GTX 1070 8GB) with an i7-6700K. Setup is Slipstream, a custom Nuitka-compiled Python engine, using a mixed-vendor Vulkan backend with Flash Attention across PCIe x1 and x2 lanes. The figure is from concurrent serving of 27B, 9B and 2B models, handling OCR and vision parsing.