PSA for DeepSeek-V4-Flash-0731 users — don't blow out your prompt cache with system role messages mid-conversation

DSv4F doesn't ship a jinja, but for distributions that do and faithfully reconstruct what DS releases in their chat template python, every system message is hoisted into the system prompt at the top -- the format has no mid-conversation system turn. So, anything you stick at the tail or mid-convo ac

中文全文翻译正在补齐，可先查看下方英文原文。

DSv4F doesn't ship a jinja, but for distributions that do and faithfully reconstruct what DS releases in their chat template python, every system message is hoisted into the system prompt at the top -- the format has no mid-conversation system turn. So, anything you stick at the tail or mid-convo ac