DeepSeek-V4-Flash Benchmarks Leak: Higher IQ Density but Questionable Pricing

teortaxesTex · x · 2026-07-31

According to leaked information, DeepSeek-V4-Flash demonstrates excellent performance across multiple benchmarks like AutomationBench. While it generates text similar to the Preview version, it offers a higher IQ density per token.

However, users noted that its pricing might be 3x that of the Luna model, sparking debates about its cost-effectiveness, especially since DeepSeek does not charge for cache writes.

Related event: DeepSeek-V4-Flash Benchmarks Impress but Face Pricing Controversy(4 posts)→

Original post →

More from Models

Models channel →