Qwen3.8 27B

on 3× NVIDIA RTX 5060 Ti 16GB · llama.cpp · 210,000 ctx

Sep 7, 2026

Throughput

38-44 t/s gen · 1300.0 tokens/s prompt processing (prefill)Prompt processing measures input prefill speed; it does not tell you the time to first token.

Quant

NVFP4 (GGUF)

KV cache

System RAM

128 GB

VRAM reported

16 GB

Summary

User reports Qwen3.8-27B-NVFP4-MTP-GGUF at 210k context on 3x RTX 5060 Ti 16GB in a Dell 5820 workstation with 128GB DDR4 RAM, with prompt processing at 1300-700 t/s and generation at 29-71 t/s, mostly 38-44. Setup is llama.cpp with a Q8 KV cache and vision enabled. The user added a third GPU but reports cooling and stability issues, and is considering a second workstation with subagents for Mixture of Agents. The user also tested Ornith 1.5 35B (MXFP) and Gemma 4 26B NVFP on 16GB VRAM. Use cases are web hosting, office documents, and prompt generation for image and video models.

View source ↗