V4.1-Flash KV cache: 1/4 HBM, 1/8 SSD storage vs prior generation. Cache-hit charges are major agent cost driver; compression cuts costs significantly.
From Sept 14, 2026 04:00 UTC, deepseek-v4-pro requests route to V4.1-Flash at V4.1-Flash rates until V4.1-Pro launches.
Peak/off-peak pricing continues; off-peak 50% of peak. New pricing effective Sept 10, 2026 04:00 UTC.
Partners include WorkBuddy passing savings to users.