DeepSeek-V4.1-Flash Pricing on SiliconFlow: Off-Peak Rates for Fast, Low-Cost Inference

SiliconFlow ·

Key Info

SiliconFlow has released pricing for DeepSeek-V4.1-Flash, now live on the platform, with cheaper rates during off-peak hours (14:00–24:00 UTC) and higher rates during peak hours (00:00–14:00 UTC).

Highlights

  • Off-peak pricing: cache reads at $0.003/M tokens, input at $0.15/M tokens, and output at $0.60/M tokens.
  • Peak pricing: cache reads at $0.006/M tokens, input at $0.30/M tokens, and output at $1.20/M tokens.
  • The pricing matches the “Flash” branding, offering a low-cost option for high-volume inference workloads.
  • Developers can optimize spend by shifting non-urgent traffic to the off-peak window.
Loading...