benchr
Pricing Guide · DeepSeek

DeepSeek V4-Flash API Pricing

Detailed token costs, context window limits, benchmark performance, and caching structures for DeepSeek V4-Flash.

Last reviewed: June 6, 2026 Sourced from official docs

DeepSeek V4-Flash Cost Breakdown

Dimension Value / Cost
Input Tokens Cost / 1M$0.140
Output Tokens Cost / 1M$0.280
Cached Input Tokens Cost / 1M$0.003
Max Context Window1,000,000 tokens
SWE-bench Verified Score79.0%
GPQA Diamond Score88.1%

Best suited for:

  • Cheapest frontier-grade inference on the market
    • \n
  • Self-hosted at low cost
    • \n
  • Volume coding tasks

← Back to all pricing guides