---
title: "DeepSeek V4.1 Flash looks contaminated with public benchmarks. You can see it in the data they..."
date: 2026-09-13
canonical: https://solmaz.io/x/2099016722322006311/
x_url: https://x.com/onusoz/status/2099016722322006311
license: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/)
---

DeepSeek V4.1 Flash looks contaminated with public benchmarks. You can see it in the data they have put in the model card, possibly Kimi K3 as well

Models that are contaminated have inverse proportionality in rank as they are tested in later benchmarks that were not present during their training  

Regardless of its intentionality, it's an amazing model. Its value comes not from higher quality output, but architectural innovation