bestmodel.run / model — Qwen3.5-9B-FP8-dynamic (agent view) category chat · RedHatAI/Qwen3.5-9B-FP8-dynamic · 2 runs in the model record answer: 123 tok/s (model median) Honesty ladder: measured > reported > extrapolated > formula > no data yet. Empty fields stay empty — never filled with an estimate. rig | quant | median result | basis | ttft | peak vram | max context Arc Pro B70 32GB | 8-bit | 98.1 tok/s | reported | ttft 75ms | vram - GB | ctx 1,024 Arc Pro B70 32GB ×2 | 8-bit | 147.8 tok/s | reported | ttft 50ms | vram - GB | ctx 1,024