Qwen 3.8 hits 45+ T/ps on M-series chips with detailed optimization guide
Adventurous_Cat_1559 · reddit · 2026-08-21
The author shares optimization experience for running Qwen 3.8 (8-bit quantized) on M-series chips, achieving a stable 45+ T/ps throughput through trial and error, setting a record for this model on the M-series. The post links to a benchmark page with detailed launch arguments, including annotations on what worked and what didn't. The author reports no performance degradation in actual coding tests compared to default settings.
More from Infra
- Nvidia's developer base hits 11 million, up ~3.7x from 3 million in 2021 — Beth_Kindig · 2026-10-16
- Muse Spark Open runs locally on a Mac M3 Ultra via llama.cpp: frontier model on desktop — pcuenq · 2026-10-16
- Muse-Spark open-weights drop: 816B-A42B MoE model gets 359GB MLX 3-bit build — ns123abc · 2026-10-16
- Meta launches Muse Spark Open, says it runs on two M5 Ultra Mac Studios with Apple's help — The Verge AI · 2026-10-16
- vLLM ships Day-0 LoRA RL for Muse Spark Open, lifting AIME from 0.56 to 0.85 — vllm_project · 2026-10-16
- DSpark draft model for Muse Spark Open proposes 7 tokens per pass with 23-42% acceptance — vllm_project · 2026-10-16