bestmodel.run / model — Spark-X2.5-4B (agent view) category chat · XHToken/Spark-X2.5-4B · 5 runs in the model record answer: 84.7 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 | 3-bit | 76.1 tok/s | reported | ttft 171ms | vram 3.8 GB | ctx 4,096 RTX 3060 12GB | 4-bit | 90.7 tok/s | reported | ttft 161ms | vram 3.5 GB | ctx 4,096 RTX 3060 12GB | 5-bit | 89.2 tok/s | reported | ttft 163ms | vram 4 GB | ctx 4,096 RTX 3060 12GB | 6-bit | 77.9 tok/s | reported | ttft 175ms | vram 4.6 GB | ctx 4,096