[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
+5
View File
@@ -21,6 +21,7 @@ from vllm.entrypoints.openai.models.serving import OpenAIServingModels
from vllm.entrypoints.serve.engine.serving import BaseServing
from vllm.entrypoints.serve.engine.typing import AnyRequest
from vllm.entrypoints.serve.utils.request_logger import RequestLogger
from vllm.exceptions import GenerationError, RetryableRequestError
from vllm.lora.request import LoRARequest
from vllm.renderers.base import BaseRenderer
from vllm.tracing import (
@@ -206,6 +207,10 @@ class PoolingBaseServing(ABC, BaseServing):
final_res_batch = [None] * num_inputs
async for i, res in ctx.result_generator:
if res.error is not None:
if res.error.retryable:
raise RetryableRequestError(res.error.message)
raise GenerationError(res.error.message)
final_res_batch[i] = res
if None in final_res_batch: