StepFun 3.7 198B (11B active) Flash
on M5 Max 128GB · llama.cpp · 65,536 ctx
Aug 28, 2026
Summary
User benchmarks prompt processing at 65,536 context, with throughput falling from 62.80 t/s at 128 context to 33.92 t/s at 65,536.
Memory peaked at ~120+ GB, and the system stayed sluggish but usable.