Why server slot's cache_prompt is false by default?

[Nekotekina]

What happened?

I have no idea whether this is intended or not but it doesn't make much sense to not use cached prompt. It makes prompt processing very slow. And it doesn't seem easy to inject explicit parameter to 3rdparty OpenAI-compatible request.

Name and Version

ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no
ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no
ggml_cuda_init: found 1 CUDA devices:
Device 0: NVIDIA GeForce RTX 3090, compute capability 8.6, VMM: yes
version: 4142 (8fd4b7f)
built with cc (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0 for x86_64-linux-gnu

What operating system are you seeing the problem on?

No response

Relevant log output

slot update_slots: id  0 | task 365 | new prompt, n_ctx_slot = 8192, n_keep = 0, n_prompt_tokens = 1957