Is Anthropic prompt caching enabled when using Bedrock as the provider?

[Defiect]

Looking at changelogs for recent LibreChat versions, I've seen that prompt caching has been developed when calling Anthropic models directly via their API (and I can see there's a toggle for it in the UI)

I was wondering if it's enabled when calling e.g. Sonnet 4 via bedrock ?
(us.anthropic.claude-sonnet-4-20250514-v1:0 is the inference profile used, thus my .env has this line:)
BEDROCK_AWS_MODELS=us.anthropic.claude-opus-4-1-20250805-v1:0

This AWS docs page implies that maybe it's on by default if LibreChat is calling the InvokeModel API - I'm not sure if we use Converse or InvokeModel though. https://docs.aws.amazon.com/bedrock/latest/userguide/prompt-caching.html

[Pfaufisch]

There's currently an open PR for that: #8271

[Defiect]

Thanks! I somehow didn't catch that