DeepSeek V4 Pro costs $1.74/1M input and $3.48/1M output tokens while V4 Flash costs $0.14/1M input and $0.28/1M output tokens, both the cheapest in their class
Chinese AI lab DeepSeek's last model release was V3.2 (and V3.2 Speciale) last December. They just dropped the first of their …
Lead Source
More
Discussion
TechSnif Coverage
DeepSeek Drops V4 Models With Rock-Bottom API Pricing
DeepSeek's new V4 Pro and V4 Flash models undercut every competitor in their respective pricing tiers.
Chinese AI lab DeepSeek just launched its V4 lineup, and the price tags are aggressive. V4 Pro clocks in at $1.74 per million input tokens and $3.48 per million output tokens. V4 Flash goes even cheaper — $0.14 per million input and $0.28 per million output.
Both models claim the title of cheapest in their class. That's not a minor distinction. It puts serious downward pressure on API pricing across the entire industry.
DeepSeek's last release was V3.2 and V3.2 Speciale back in December. The jump to V4 represents the company's next major generation, and pricing appears to be the headline weapon.
The Flash tier especially stands out. At 14 cents per million input tokens, it's practically spare change for developers building token-heavy applications. The AI pricing war just got another round of ammunition.