Skip to content
lifeidle / qwen3.8-27b-5090-laptop-24gb-coder390-ninferView on GitHub
Qwen3.8-27B on RTX 5090 Laptop 24GB — TWIN-TURBO NVFP4 (native llama.cpp) vs Coder390 Q3 LynnStyle (.ninfer + NInfer engine in WSL2/Docker): 86 tok/s decode, 262K context, C4 aggregate 228 tok/s. Every number a controlled measurement.
☆24Oct 16, 2026Updated last month

Alternatives and similar repositories for qwen3.8-27b-5090-laptop-24gb-coder390-ninfer

Users that are interested in qwen3.8-27b-5090-laptop-24gb-coder390-ninfer are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.

Sorting:

Are these results useful?