Misc. bug: Llama-Server is missing --Prompt-Cache from Llama-CLI

[ColumbusAI]

Name and Version

Llama-server is missing prompt caching from llama-cli.

Here are the parameters in Llama-cli:

--prompt-cache FNAME                    file to cache prompt state for faster startup (default: none)
--prompt-cache-all                      if specified, saves user input and generations to cache as well
--prompt-cache-ro                       if specified, uses the prompt cache but does not update it

Figured it might be a bug as, for the most part, I've seen feature parity across these binaries. May we please have this addressed?

Thank you all.

Operating systems

Linux

Which llama.cpp modules do you know to be affected?

llama-server

Command line

Problem description & steps to reproduce

See above

First Bad Commit

No response

Relevant log output

[mdenomme24]

llama-server has the capability of running multiple slots each with their own cache, the functionality you are looking for is implemented with a server API endpoint
POST /slots/{id_slot}?action=save: Save the prompt cache of the specified slot to a file.
POST /slots/{id_slot}?action=restore: Restore the prompt cache of the specified slot from a file.

example:

start server:
llama-server -m ./models/yourmodel.gguf --slot-save-path ./kvcache/

save cache:
curl -X POST "http://127.0.0.1:8080/slots/0?action=save" -H "Content-Type: application/json" -d '{"filename": "my_cache.bin"}'

restore cache:
curl -X POST "http://127.0.0.1:8080/slots/0?action=restore" -H "Content-Type: application/json" -d '{"filename": "my_cache.bin"}'

[ColumbusAI]

Got it, thank you @mdenomme24 ! Case closed.