Qwen 3.8 27B Shows Stable Local Performance

Developers tested Qwen 3.8 27B on RTX 3090 and 4090 GPUs, showing stable performance at 37-73 tok/s with 4bit quantization and 128K context support.

2026-08-16 ~ 2026-08-17 · 2 related posts

• Qwen 3.8 27b with DeepSeek Harness: 10h stable, 37-60 tok/s on RTX 3090 — cviperr33 · 2026-08-16
• Qwen 3.8 27B Local Coding: 73 tok/s at 128K Context on RTX 4090 — julianharris · 2026-08-17