llamaperf

Qwen3.8 125B (6B active) Flash-Next

on M4 Max 48GB · Strata

Tone: positiveApple hardware
Nov 8, 2026
Throughput
25.0 t/s gen
Quant
IQ3_XXS (GGUF)
System RAM
48 GB

Use cases

coding

Summary

User reports Qwen3.8 Flash-Next at 25 t/s on an M4 Max 48GB, running in Strata with IQ3_XXS weights. Setup is the Strata GUI app on a Mac Studio; the user says it runs smoothly for coding and has already built a gamma spectrometry data analysis app with it. The post is a question about generating images locally with the same LLM, not a benchmark; the user wants to avoid cloud services.

Compare this run

Medians of plain runs: one device, one request, no speculative decoding, the whole model in the device's memory. The report above is one setup, so read it beside them and not as one of them.

Qwen3.8 125B · 6B active at 4-bit by hardware

Bars are medians on one scale, 0 to 232.0 tokens/s. Each dot is one run and opens its report. Press ▶ to watch text arrive at a row's speed.

Models most run on the M4 Max 48GB

Bars are medians on one scale, 0 to 85.0 tokens/s. Each dot is one run and opens its report. Press ▶ to watch text arrive at a row's speed.

Everything reported on the M4 Max 48GB →Every Qwen3.8 report →Estimate Qwen3.8 on this card →