[Bugfix][Pooling] Handle multimodal cache misses without crashing (#58975)

Signed-off-by: linchaoying6 <linchaoying6@jd.com>
Co-authored-by: linchaoying6 <linchaoying6@jd.com>
Co-authored-by: OpenAI Codex <codex@openai.com>
This commit is contained in:
Chaoying
2026-10-02 10:14:33 -04:00
committed by GitHub
co-authored by linchaoying6 OpenAI Codex
parent 443dccd3c9
commit 592c6f3fbc
21 changed files with 525 additions and 17 deletions
+8
View File
@@ -130,6 +130,14 @@ class GenerationError(VLLMServerError):
self.status_code = HTTPStatus.INTERNAL_SERVER_ERROR
class RetryableRequestError(GenerationError):
"""A transient request-level failure that is safe for clients to retry."""
def __init__(self, message: str = "Request failed; please retry"):
super().__init__(message)
self.status_code = HTTPStatus.SERVICE_UNAVAILABLE
class GracefulHTTPError(VLLMError):
"""Exception that should be translated into an HTTP error response.