phoenix/litellm-mirror
0
0
Fork 1
mirror of https://github.com/BerriAI/litellm.git synced 2025-04-26 11:14:04 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

[UI] Allow setting prompt cache_control_injection_points (#10000)

Browse source 
* test_anthropic_cache_control_hook_system_message

* test_anthropic_cache_control_hook.py

* should_run_prompt_management_hooks

* fix should_run_prompt_management_hooks

* test_anthropic_cache_control_hook_specific_index

* fix test

* fix linting errors

* ChatCompletionCachedContent

* initial commit for cache control

* fixes ui design

* fix inserting cache_control_injection_points

* fix entering cache control points

* fixes for using cache control on ui + backend

* update cache control settings on edit model page

* fix init custom logger compatible class

* fix linting errors

* fix linting errors

* fix get_chat_completion_prompt
This commit is contained in:
Ishaan Jaff 2025-04-14 21:17:42 -07:00 • committed by GitHub
parent 990fda294b
commit 1a9a772dab
16 changed files with 358 additions and 39 deletions
Download patch file Download diff file Expand all files Collapse all files

2
litellm/proxy/custom_prompt_management.py
View file

@ -12,7 +12,7 @@ class X42PromptManagement(CustomPromptManagement):
model: str,
messages: List[AllMessageValues],
non_default_params: dict,
prompt_id: str,
prompt_id: Optional[str],
prompt_variables: Optional[dict],
dynamic_callback_params: StandardCallbackDynamicParams,
) -> Tuple[str, List[AllMessageValues], dict]: