DeepSeek V4.1 Flash Popped All 11 Vulnerable Targets in Enclave's Offensive Security Benchmark for $5.14

Enclave (model release corroborated by DeepSeek) / Hacker News (167pts, 66 comments)·medium signalXBlueskyLinkedInCopy link

Enclave published results on September 16 in which DeepSeek V4.1 Flash gained code execution on all 11 vulnerable targets including Grafana, Jenkins and Nextcloud, while all four correctly patched targets held. It found the six planned attack routes plus five unexpected ones. Total spend was $5.14 including failed attempts ($4.65 for accepted runs), achieved because 266.2 million of 268.3 million input tokens were cache reads, which is the practical consequence of the KV cache compression DeepSeek shipped with V4.1-Flash on September 10 (one quarter the HBM and one eighth the SSD of the prior generation).