Any plans to add prompt caching for Anthropic?

[tavurth]

Why?

When you send a request with Prompt Caching enabled:

The system checks if the prompt prefix is already cached from a recent query. If found, it uses the cached version, reducing processing time and costs. Otherwise, it processes the full prompt and caches the prefix for future use.

• Cache write tokens are 25% more expensive than base input tokens
• Cache read tokens are 90% cheaper than base input tokens
• Regular input and output tokens are priced at standard rates

This is especially useful for:

• Prompts with many examples
• Large amounts of context or background information
• Repetitive tasks with consistent instructions
• Long multi-turn conversations

Seems like it can be super useful for large codebases where only a small section needs to be updated each time. Can basically send it all over and it'll be cached each time.

https://docs.anthropic.com/en/docs/build-with-claude/prompt-caching#how-prompt-caching-works

[CestDiego]

this would be perfect for having a similar feature to projects from Anthropic as well

[karthink]

Yup, will add it soon.