PromptTemplate with Prompt Caching for ChatAnthropic

[yadgire7]

Checked other resources

• I added a very descriptive title to this question.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.

Commit to Help

• I commit to help with one of those options 👆

Example Code

return ChatPromptTemplate.from_messages([
        SystemMessage(content=[
            {
                "text": cached_prompt,
                "type": "text",
                "cache_control": {"type": "ephemeral"},
            }
        ]),
        HumanMessage(content=f"Here is the comment to analyze: {comment}")
    ])

Description

• I want to use prompt caching to reduce latency.

• My earlier approach used PromptTemplate with context to extract file name from a large description of the folder structure and file structures (the the complete directory tree with all the metadata about contents, size, creation date, etc). And it works well.

• Now, I want to cache this large context as it does not change much over time.

• I implemented the above code snippet by providing the detailed prompt as a system prompt. But it hallucinates badly. It is unable to understand the natural language question given as a user message.

• Is it possible to use prompt caching with PromptTemplate?
  Thank you

System Info

langchain 0.3.0
langchain-anthropic 0.2.0
langchain-community 0.3.0
langchain-core 0.3.0
langchain-experimental 0.3.0
langchain-openai 0.2.0
langchain-text-splitters 0.3.0