StepFun / Step 3.7 Flash
Step 3.7 Flash FP8 quantization
Estimated until measured during engineering
Before → After
BasisversionVerifizierter Checkpoint
198.000B
198.000B
quantizationFP8
Quality retention target
Quality retention target
ProduktionsartefaktArtefakt
Lineage- und Benchmarkbericht
Lineage- und Benchmarkbericht
GPU-Zielsetzung
Gewichte und KV-CacheEstimated until measured during engineering
Estimated VRAMMeasured and certified results are delivered with the project.
Runtime-Paketierungtransformers