DeepSeek V4 Flash

by DeepSeek model

The cheaper half of DeepSeek's current pair: the same 1M-token context window and 384K max output as V4 Pro, at $0.14 per million input tokens on a cache miss and $0.28 per million output — about a third of Pro's price. Cache hits drop the input rate to $0.0028. From 16 August 2026 off-peak hours bill at half the peak rate, peak being 01:00-04:00 and 06:00-10:00 UTC.

Visit site

Good at

• Chat 4 / 5
• Code Generation 4 / 5
• Text Generation 4 / 5