bestmodel.run / model — Qwen3.8-27B-QUASAR-NVFP4 (agent view) category chat · QUASAR-QAT/Qwen3.8-27B-QUASAR-NVFP4 · 2 runs in the model record answer: 263.7 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 5070 Ti 16GB ×2 | 4-bit | 100.1 tok/s | reported | ttft -ms | vram 29 GB | ctx 2,048 Tesla V100 32GB ×4 | 4-bit | 427.3 tok/s | reported | ttft 315ms | vram 117.4 GB | ctx 262,144