DeepSeek V4.1 Flash Launches on Requesty: Up to 57% Cheaper, Now with Native Image Input

Requesty ·

Key Info

DeepSeek V4.1 Flash, the latest small model in DeepSeek's new architecture family, is now available on Requesty with lower per-token prices — input $0.30/M, cache hit $0.006/M, output $1.20/M — plus native image understanding and a 1M-token context window.

Highlights

  • New model brings 32% cheaper input, 57% cheaper cache hits, and 9% cheaper output compared to DeepSeek v4 Flash pricing.
  • Adds native visual understanding and a 1M-token context, while emphasizing faster inference and higher throughput for production workloads.
Loading...