fix: use cache_salt for gpt-oss (#23186)

Signed-off-by: Marko Rosenmueller <5467316+dr75@users.noreply.github.com>
This commit is contained in:
Marko Rosenmueller
2025-08-19 18:12:25 +00:00
committed by GitHub
parent b94faf9d50
commit 80141bbf2f
3 changed files with 13 additions and 1 deletions
+5
View File
@@ -1483,4 +1483,9 @@ class OpenAIServingChat(OpenAIServing):
# Render prompt token ids.
prompt_token_ids = render_for_completion(messages)
engine_prompt = EngineTokensPrompt(prompt_token_ids=prompt_token_ids)
# Add cache_salt if provided in the request
if request.cache_salt is not None:
engine_prompt["cache_salt"] = request.cache_salt
return messages, [prompt_token_ids], [engine_prompt]
@@ -408,6 +408,11 @@ class OpenAIServingResponses(OpenAIServing):
request, prev_response)
prompt_token_ids = render_for_completion(messages)
engine_prompt = EngineTokensPrompt(prompt_token_ids=prompt_token_ids)
# Add cache_salt if provided in the request
if request.cache_salt is not None:
engine_prompt["cache_salt"] = request.cache_salt
return messages, [prompt_token_ids], [engine_prompt]
async def responses_full_generator(