Model results · DeepSeek

DeepSeek V4-Flash

Report13
DeepSeek V4 Pro vs V4-Flash benchmark
DeepSeek V4 Pro ties V4-Flash at 87.0% pass@1 on VulcanBench v3. Pro uses 55% fewer tokens and runs 38% faster, but its higher token price makes the sweep 29% more expensive.
2026-08-09

How would DeepSeek V4-Flash do on your stack?

Public results answer "which model is stronger on this suite." Whether DeepSeek V4-Flash belongs in your routing table depends on your languages, your codebases, and your task mix — that gets measured, not guessed. Grading rules are in the methodology.