[Frontend] Add detokenization streaming derender for disaggregated serving (#47301)

Signed-off-by: Martin Hickey <martin.hickey@ie.ibm.com>
This commit is contained in:
Martin Hickey
2026-07-29 18:01:20 -07:00
committed by GitHub
parent 451227cb3f
commit 0a31372e5f
6 changed files with 1377 additions and 32 deletions
+4
View File
@@ -17,7 +17,9 @@ from vllm.entrypoints.openai.completion.protocol import (
from vllm.entrypoints.openai.responses.protocol import ResponsesRequest
from vllm.entrypoints.scale_out.token_in_token_out.protocol import (
DerenderChatRequest,
DerenderChatStreamRequest,
DerenderCompletionRequest,
DerenderCompletionStreamRequest,
GenerateRequest,
GenerateResponse,
)
@@ -54,6 +56,7 @@ CompletionLikeRequest: TypeAlias = (
| TokenizeCompletionRequest
| DetokenizeRequest
| DerenderCompletionRequest
| DerenderCompletionStreamRequest
)
ChatLikeRequest: TypeAlias = (
@@ -61,6 +64,7 @@ ChatLikeRequest: TypeAlias = (
| BatchChatCompletionRequest
| TokenizeChatRequest
| DerenderChatRequest
| DerenderChatStreamRequest
)
SpeechToTextRequest: TypeAlias = TranscriptionRequest | TranslationRequest