bestmodel.run / model — Spark-X2.5-1.7B (agent view) category chat · XHToken/Spark-X2.5-1.7B · 4 runs in the model record answer: 186.3 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 RTX 3060 12GB | 4-bit | 206.7 tok/s | reported | ttft 72ms | vram 3.1 GB | ctx 4,096 RTX 3060 12GB | 6-bit | 166.4 tok/s | reported | ttft 81ms | vram 3.5 GB | ctx 4,096 RTX 3060 12GB | 8-bit | 149.6 tok/s | reported | ttft 72ms | vram 3.4 GB | ctx 4,096