o
    پi{0                     @   s0  d Z ddlmZ ddlmZ ddlmZ ddlmZ	 ddl
mZ eejjddd	d
d e	 Zd	dlmZ e dZe Zeee eede ejsde_ded _ded _ded _ded _ded _ded _ded _ded _ded _ded _ded _ded _ded _ded _ded  _d!ed  _d"ed# _d$ed# _d%ed& _d'ed& _d(ed _d)ed _d*ed _d+ed _d,ed- _d.ed- _d/ed0 _d1ed0 _d2ed3 _d4ed3 _d5ed6 _d7ed6 _d8ed9 _d:ed9 _d;ed< _d=ed< _d>ed? _d@ed? _dAedB _dCedB _dDedE _dFedE _dGedH _dIedH _dJedK _dLedK _dMedN _dOedN _dPedQ _dRedQ _dSedT _dUedT _dVedW _dXedW _dYedZ _d[edZ _d\ed] _d^ed] _d_ed` _daed` _dbedc _ddedc _deedf _dgedf _dS dS )hzGenerated protocol buffer code.    )
descriptor)descriptor_pool)runtime_version)symbol_database)builder          zvllm_engine.proto)
common_pb2sD  
vllm_engine.protovllm.grpc.enginecommon.proto"
SamplingParams
temperature (H
top_p (
top_k (
min_p (
frequency_penalty (
presence_penalty (
repetition_penalty (

max_tokens (H

min_tokens	 (
stop
 (	
stop_token_ids (
skip_special_tokens (%
spaces_between_special_tokens (

ignore_eos (	
n (
logprobs (H
prompt_logprobs (H
seed (H"
include_stop_str_in_output (C

logit_bias (2/.vllm.grpc.engine.SamplingParams.LogitBiasEntry#
truncate_prompt_tokens (H
json_schema (	H 
regex (	H 
grammar (	H 
structural_tag (	H 
json_object (H 4
choice (2".vllm.grpc.engine.ChoiceConstraintH 0
LogitBiasEntry
key (
value (:8B

constraintB
_temperatureB
_max_tokensB
	_logprobsB
_prompt_logprobsB
_seedB
_truncate_prompt_tokens"#
ChoiceConstraint
choices (	":
TokenizedInput
original_text (	
	input_ids ("8

TensorData
data (
shape (
dtype (	"2
PlaceholderRange
offset (
length ("
MultimodalInputs2
pixel_values (2.vllm.grpc.engine.TensorData\
model_specific_tensors (2<.vllm.grpc.engine.MultimodalInputs.ModelSpecificTensorsEntry
im_token_id (H ;
mm_placeholders (2".vllm.grpc.engine.PlaceholderRange
	mm_hashes (	
batched_keys (	C
	flat_keys (20.vllm.grpc.engine.MultimodalInputs.FlatKeysEntry
keep_on_cpu_keys	 (	Y
ModelSpecificTensorsEntry
key (	+
value (2.vllm.grpc.engine.TensorData:8/
FlatKeysEntry
key (	
value (	:8B
_im_token_idJ"
GenerateRequest

request_id (	5
	tokenized (2 .vllm.grpc.engine.TokenizedInputH 
text (	H 9
sampling_params (2 .vllm.grpc.engine.SamplingParams
stream (>
kv_transfer_params (2".vllm.grpc.engine.KvTransferParams5
	mm_inputs (2".vllm.grpc.engine.MultimodalInputsB
input"<
KvTransferParams
remote_host (	
remote_port ("p
OutputLogProbs
token_logprobs (
	token_ids (3
top_logprobs (2.vllm.grpc.engine.TopLogProbs"
InputLogProbs;
token_logprobs (2#.vllm.grpc.engine.InputTokenLogProb
	token_ids (3
top_logprobs (2.vllm.grpc.engine.TopLogProbs"1
InputTokenLogProb
value (H B
_value"0
TopLogProbs
values (
	token_ids ("
GenerateResponse6
chunk (2%.vllm.grpc.engine.GenerateStreamChunkH 6
complete (2".vllm.grpc.engine.GenerateCompleteH B

response"
GenerateStreamChunk
	token_ids (
prompt_tokens (
completion_tokens (
cached_tokens (9
output_logprobs (2 .vllm.grpc.engine.OutputLogProbs7
input_logprobs (2.vllm.grpc.engine.InputLogProbs
index ("
GenerateComplete

output_ids (
finish_reason (	
prompt_tokens (
completion_tokens (
cached_tokens (9
output_logprobs (2 .vllm.grpc.engine.OutputLogProbs7
input_logprobs (2.vllm.grpc.engine.InputLogProbs
index (>
kv_transfer_params	 (2".vllm.grpc.engine.KvTransferParams
matched_token_id
 (H 
matched_stop_str (	H B
matched_stop"W
EmbedRequest

request_id (	3
	tokenized (2 .vllm.grpc.engine.TokenizedInput"P
EmbedResponse
	embedding (
prompt_tokens (
embedding_dim ("
HealthCheckRequest"7
HealthCheckResponse
healthy (
message (	"#
AbortRequest
request_ids (	"
AbortResponse"
GetModelInfoRequest"
GetModelInfoResponse

model_path (	
is_generation (
max_context_length (

vocab_size (
supports_vision (
served_model_name (	"
GetServerInfoRequest"
GetServerInfoResponse
active_requests (
	is_paused (
last_receive_timestamp (
uptime_seconds (
server_type (	
kv_connector (	
kv_role (	2

VllmEngineS
Generate!.vllm.grpc.engine.GenerateRequest".vllm.grpc.engine.GenerateResponse0H
Embed.vllm.grpc.engine.EmbedRequest.vllm.grpc.engine.EmbedResponseZ
HealthCheck$.vllm.grpc.engine.HealthCheckRequest%.vllm.grpc.engine.HealthCheckResponseH
Abort.vllm.grpc.engine.AbortRequest.vllm.grpc.engine.AbortResponse]
GetModelInfo%.vllm.grpc.engine.GetModelInfoRequest&.vllm.grpc.engine.GetModelInfoResponse`
GetServerInfo&.vllm.grpc.engine.GetServerInfoRequest'.vllm.grpc.engine.GetServerInfoResponseZ
GetTokenizer$.smg.grpc.common.GetTokenizerRequest".smg.grpc.common.GetTokenizerChunk0_
SubscribeKvEvents).smg.grpc.common.SubscribeKvEventsRequest.smg.grpc.common.KvEventBatch0bproto3vllm_engine_pb2N_SAMPLINGPARAMS_LOGITBIASENTRYs   8+_MULTIMODALINPUTS_MODELSPECIFICTENSORSENTRY_MULTIMODALINPUTS_FLATKEYSENTRY6   _SAMPLINGPARAMSi  i  i>  i  _CHOICECONSTRAINTi  i  _TOKENIZEDINPUTi  i  _TENSORDATAiK  iM  _PLACEHOLDERRANGEi  i  _MULTIMODALINPUTSi  i  i`  ib  i  i  _GENERATEREQUESTi  i  _KVTRANSFERPARAMSi  i!  _OUTPUTLOGPROBSi  i  _INPUTLOGPROBSi(	  i*	  _INPUTTOKENLOGPROBi[	  i]	  _TOPLOGPROBSi	  i	  _GENERATERESPONSEi
  i!
  _GENERATESTREAMCHUNKi  i  _GENERATECOMPLETEi  i  _EMBEDREQUESTi  i  _EMBEDRESPONSEiT  iV  _HEALTHCHECKREQUESTij  il  _HEALTHCHECKRESPONSEi  i  _ABORTREQUESTi  i  _ABORTRESPONSEi  i  _GETMODELINFOREQUESTi  i  _GETMODELINFORESPONSEi  i  _GETSERVERINFOREQUESTi  i  _GETSERVERINFORESPONSEij  im  _VLLMENGINEi<  ) __doc__google.protobufr   _descriptorr   _descriptor_poolr   _runtime_versionr   _symbol_databasegoogle.protobuf.internalr   _builderValidateProtobufRuntimeVersionDomainPUBLICDefault_sym_dbr
   r   common__pb2AddSerializedFile
DESCRIPTORglobals_globalsBuildMessageAndEnumDescriptorsBuildTopDescriptorsAndMessages_USE_C_DESCRIPTORS_loaded_options_serialized_options_serialized_start_serialized_end rD   rD   \/home/ubuntu/.local/lib/python3.10/site-packages/smg_grpc_proto/generated/vllm_engine_pb2.py<module>   s   































































