Regime
Loading regime
Crossover
Monthly cost by tokens per day
Self-host
Hosted API
Assumptions defaults
Declared escalation
Close call analysis
This composition sits near the boundary, so RunWhere can hand it to the opt-in live path.
Provenance
Sources behind this artifact
cited: Hugging Face Hub model metadata cited: OpenRouter model catalog snapshot measured: OpenRouter endpoint throughput snapshot cited: Cloud GPU pricing snapshot calculated: memory bandwidth divided by active parameter bytes, adjusted for quantization calculated: GPU hours at 60% utilization plus 0.1 FTE labor calculated: 1.5x regime threshold
Artifact deepseek-v3-2--gcp--v1.0.0-f96c2e3b - ruleset 1.0.0