Every "other" provider other than deepseek on openrouter (or elsewhere) is charging at the very least 10x for cached input tokens... and in agentic coding, that is where the 90%+ of the costs from from... that means with these providers, your costs will jump by 10x or more.

https://openrouter.ai/deepseek/deepseek-v4-flash-0731#provid...

Sort by cache read.