Skip to content

Commit 3c0b84d

Browse files
Google APIscopybara-github
Google APIs
authored and
copybara-github
committed
feat: A new message RoutingConfig is added
feat: A new field `seed` is added to message `.google.cloud.aiplatform.v1beta1.GenerationConfig` feat: A new field `routing_config` is added to message `.google.cloud.aiplatform.v1beta1.GenerationConfig` feat: A new field `avg_logprobs` is added to message `.google.cloud.aiplatform.v1beta1.Candidate` feat: A new field `satisfies_pzs` is added to message `.google.cloud.aiplatform.v1beta1.CustomJob` feat: A new field `satisfies_pzi` is added to message `.google.cloud.aiplatform.v1beta1.CustomJob` feat: A new field `satisfies_pzs` is added to message `.google.cloud.aiplatform.v1beta1.EntityType` feat: A new field `satisfies_pzi` is added to message `.google.cloud.aiplatform.v1beta1.EntityType` feat: A new message `TimeSeries` is added feat: A new field `time_series` is added to message `.google.cloud.aiplatform.v1beta1.FeatureGroup` feat: A new field `satisfies_pzs` is added to message `.google.cloud.aiplatform.v1beta1.FeatureOnlineStore` feat: A new field `satisfies_pzi` is added to message `.google.cloud.aiplatform.v1beta1.FeatureOnlineStore` feat: A new field `satisfies_pzs` is added to message `.google.cloud.aiplatform.v1beta1.FeatureView` feat: A new field `satisfies_pzi` is added to message `.google.cloud.aiplatform.v1beta1.FeatureView` feat: A new field `satisfies_pzs` is added to message `.google.cloud.aiplatform.v1beta1.FeatureViewSync` feat: A new field `satisfies_pzi` is added to message `.google.cloud.aiplatform.v1beta1.FeatureViewSync` feat: A new field `satisfies_pzs` is added to message `.google.cloud.aiplatform.v1beta1.Featurestore` feat: A new field `satisfies_pzi` is added to message `.google.cloud.aiplatform.v1beta1.Featurestore` feat: A new field `satisfies_pzs` is added to message `.google.cloud.aiplatform.v1beta1.HyperparameterTuningJob` feat: A new field `satisfies_pzi` is added to message `.google.cloud.aiplatform.v1beta1.HyperparameterTuningJob` feat: A new field `satisfies_pzs` is added to message `.google.cloud.aiplatform.v1beta1.Index` feat: A new field `satisfies_pzi` is added to message `.google.cloud.aiplatform.v1beta1.Index` feat: A new field `satisfies_pzs` is added to message `.google.cloud.aiplatform.v1beta1.IndexEndpoint` feat: A new field `satisfies_pzi` is added to message `.google.cloud.aiplatform.v1beta1.IndexEndpoint` feat: A new field `satisfies_pzs` is added to message `.google.cloud.aiplatform.v1beta1.ModelDeploymentMonitoringJob` feat: A new field `satisfies_pzi` is added to message `.google.cloud.aiplatform.v1beta1.ModelDeploymentMonitoringJob` feat: A new field `hugging_face_token` is added to message `.google.cloud.aiplatform.v1beta1.GetPublisherModelRequest` feat: A new field `satisfies_pzs` is added to message `.google.cloud.aiplatform.v1beta1.NasJob` feat: A new field `satisfies_pzi` is added to message `.google.cloud.aiplatform.v1beta1.NasJob` feat: A new field `encryption_spec` is added to message `.google.cloud.aiplatform.v1beta1.NotebookExecutionJob` feat: A new field `sample_request` is added to message `.google.cloud.aiplatform.v1beta1.PublisherModel` feat: A new field `service_attachment` is added to message `.google.cloud.aiplatform.v1beta1.PrivateServiceConnectConfig` feat: A new field `total_truncated_example_count` is added to message `.google.cloud.aiplatform.v1beta1.SupervisedTuningDataStats` feat: A new field `truncated_example_indices` is added to message `.google.cloud.aiplatform.v1beta1.SupervisedTuningDataStats` fix!: An existing message `GroundingAttribution` is removed fix!: An existing field `grounding_attributions` is removed from message `.google.cloud.aiplatform.v1beta1.GroundingMetadata` fix!: An existing field `disable_attribution` is removed from message `.google.cloud.aiplatform.v1beta1.GoogleSearchRetrieval` docs: A comment for enum value `STOP` in enum `FinishReason` is changed docs: A comment for enum value `MAX_TOKENS` in enum `FinishReason` is changed docs: A comment for enum value `SAFETY` in enum `FinishReason` is changed docs: A comment for enum value `RECITATION` in enum `FinishReason` is changed docs: A comment for enum value `OTHER` in enum `FinishReason` is changed docs: A comment for enum value `BLOCKLIST` in enum `FinishReason` is changed docs: A comment for enum value `PROHIBITED_CONTENT` in enum `FinishReason` is changed docs: A comment for enum value `SPII` in enum `FinishReason` is changed docs: A comment for enum `Strategy` is changed docs: A comment for enum value `STRATEGY_UNSPECIFIED` in enum `Strategy` is changed docs: A comment for enum value `AUTO` in enum `Mode` is changed docs: A comment for field `distance` in message `.google.cloud.aiplatform.v1beta1.RagContexts` is changed PiperOrigin-RevId: 661167484
1 parent a353d25 commit 3c0b84d

19 files changed

Lines changed: 178 additions & 69 deletions

‎google/cloud/aiplatform/v1beta1/content.proto‎

Lines changed: 66 additions & 58 deletions
Original file line numberDiff line numberDiff line change
@@ -137,6 +137,48 @@ message VideoMetadata {
137137

138138
// Generation config.
139139
message GenerationConfig {
140+
// The configuration for routing the request to a specific model.
141+
message RoutingConfig {
142+
// When automated routing is specified, the routing will be determined by
143+
// the pretrained routing model and customer provided model routing
144+
// preference.
145+
message AutoRoutingMode {
146+
// The model routing preference.
147+
enum ModelRoutingPreference {
148+
// Unspecified model routing preference.
149+
UNKNOWN = 0;
150+
151+
// Prefer higher quality over low cost.
152+
PRIORITIZE_QUALITY = 1;
153+
154+
// Balanced model routing preference.
155+
BALANCED = 2;
156+
157+
// Prefer lower cost over higher quality.
158+
PRIORITIZE_COST = 3;
159+
}
160+
161+
// The model routing preference.
162+
optional ModelRoutingPreference model_routing_preference = 1;
163+
}
164+
165+
// When manual routing is set, the specified model will be used directly.
166+
message ManualRoutingMode {
167+
// The model name to use. Only the public LLM models are accepted. e.g.
168+
// 'gemini-1.5-pro-001'.
169+
optional string model_name = 1;
170+
}
171+
172+
// Routing mode.
173+
oneof routing_config {
174+
// Automated routing.
175+
AutoRoutingMode auto_mode = 1;
176+
177+
// Manual routing.
178+
ManualRoutingMode manual_mode = 2;
179+
}
180+
}
181+
140182
// Optional. Controls the randomness of predictions.
141183
optional float temperature = 1 [(google.api.field_behavior) = OPTIONAL];
142184

@@ -161,6 +203,9 @@ message GenerationConfig {
161203
// Optional. Frequency penalties.
162204
optional float frequency_penalty = 9 [(google.api.field_behavior) = OPTIONAL];
163205

206+
// Optional. Seed.
207+
optional int32 seed = 12 [(google.api.field_behavior) = OPTIONAL];
208+
164209
// Optional. Output response mimetype of the generated candidate text.
165210
// Supported mimetype:
166211
// - `text/plain`: (default) Text output.
@@ -178,6 +223,10 @@ message GenerationConfig {
178223
// Compatible mimetypes:
179224
// `application/json`: Schema for JSON response.
180225
optional Schema response_schema = 16 [(google.api.field_behavior) = OPTIONAL];
226+
227+
// Optional. Routing configuration.
228+
optional RoutingConfig routing_config = 17
229+
[(google.api.field_behavior) = OPTIONAL];
181230
}
182231

183232
// Safety settings.
@@ -317,34 +366,34 @@ message Candidate {
317366
// The finish reason is unspecified.
318367
FINISH_REASON_UNSPECIFIED = 0;
319368

320-
// Natural stop point of the model or provided stop sequence.
369+
// Token generation reached a natural stopping point or a configured stop
370+
// sequence.
321371
STOP = 1;
322372

323-
// The maximum number of tokens as specified in the request was reached.
373+
// Token generation reached the configured maximum output tokens.
324374
MAX_TOKENS = 2;
325375

326-
// The token generation was stopped as the response was flagged for safety
327-
// reasons. NOTE: When streaming the Candidate.content will be empty if
328-
// content filters blocked the output.
376+
// Token generation stopped because the content potentially contains safety
377+
// violations. NOTE: When streaming,
378+
// [content][google.cloud.aiplatform.v1beta1.Candidate.content] is empty if
379+
// content filters blocks the output.
329380
SAFETY = 3;
330381

331-
// The token generation was stopped as the response was flagged for
332-
// unauthorized citations.
382+
// Token generation stopped because the content potentially contains
383+
// copyright violations.
333384
RECITATION = 4;
334385

335-
// All other reasons that stopped the token generation
386+
// All other reasons that stopped the token generation.
336387
OTHER = 5;
337388

338-
// The token generation was stopped as the response was flagged for the
339-
// terms which are included from the terminology blocklist.
389+
// Token generation stopped because the content contains forbidden terms.
340390
BLOCKLIST = 6;
341391

342-
// The token generation was stopped as the response was flagged for
343-
// the prohibited contents.
392+
// Token generation stopped for potentially containing prohibited content.
344393
PROHIBITED_CONTENT = 7;
345394

346-
// The token generation was stopped as the response was flagged for
347-
// Sensitive Personally Identifiable Information (SPII) contents.
395+
// Token generation stopped because the content potentially contains
396+
// Sensitive Personally Identifiable Information (SPII).
348397
SPII = 8;
349398

350399
// The function call generated by the model is invalid.
@@ -357,6 +406,9 @@ message Candidate {
357406
// Output only. Content parts of the candidate.
358407
Content content = 2 [(google.api.field_behavior) = OUTPUT_ONLY];
359408

409+
// Output only. Average log probability score of the candidate.
410+
double avg_logprobs = 9 [(google.api.field_behavior) = OUTPUT_ONLY];
411+
360412
// Output only. The reason why the model stopped generating tokens.
361413
// If empty, the model has not stopped generating the tokens.
362414
FinishReason finish_reason = 3 [(google.api.field_behavior) = OUTPUT_ONLY];
@@ -445,46 +497,6 @@ message GroundingSupport {
445497
repeated float confidence_scores = 3;
446498
}
447499

448-
// Grounding attribution.
449-
message GroundingAttribution {
450-
// Attribution from the web.
451-
message Web {
452-
// Output only. URI reference of the attribution.
453-
string uri = 1 [(google.api.field_behavior) = OUTPUT_ONLY];
454-
455-
// Output only. Title of the attribution.
456-
string title = 2 [(google.api.field_behavior) = OUTPUT_ONLY];
457-
}
458-
459-
// Attribution from context retrieved by the retrieval tools.
460-
message RetrievedContext {
461-
// Output only. URI reference of the attribution.
462-
string uri = 1 [(google.api.field_behavior) = OUTPUT_ONLY];
463-
464-
// Output only. Title of the attribution.
465-
string title = 2 [(google.api.field_behavior) = OUTPUT_ONLY];
466-
}
467-
468-
oneof reference {
469-
// Optional. Attribution from the web.
470-
Web web = 3 [(google.api.field_behavior) = OPTIONAL];
471-
472-
// Optional. Attribution from context retrieved by the retrieval tools.
473-
RetrievedContext retrieved_context = 4
474-
[(google.api.field_behavior) = OPTIONAL];
475-
}
476-
477-
// Output only. Segment of the content this attribution belongs to.
478-
Segment segment = 1 [(google.api.field_behavior) = OUTPUT_ONLY];
479-
480-
// Optional. Output only. Confidence score of the attribution. Ranges from 0
481-
// to 1. 1 is the most confident.
482-
optional float confidence_score = 2 [
483-
(google.api.field_behavior) = OPTIONAL,
484-
(google.api.field_behavior) = OUTPUT_ONLY
485-
];
486-
}
487-
488500
// Metadata returned to client when grounding is enabled.
489501
message GroundingMetadata {
490502
// Optional. Web search queries for the following-up web search.
@@ -499,10 +511,6 @@ message GroundingMetadata {
499511
repeated string retrieval_queries = 3
500512
[(google.api.field_behavior) = OPTIONAL];
501513

502-
// Optional. List of grounding attributions.
503-
repeated GroundingAttribution grounding_attributions = 2
504-
[(google.api.field_behavior) = OPTIONAL];
505-
506514
// List of supporting references retrieved from specified grounding source.
507515
repeated GroundingChunk grounding_chunks = 5;
508516

‎google/cloud/aiplatform/v1beta1/custom_job.proto‎

Lines changed: 10 additions & 4 deletions
Original file line numberDiff line numberDiff line change
@@ -109,6 +109,12 @@ message CustomJob {
109109
// The values are the URIs for each node's interactive shell.
110110
map<string, string> web_access_uris = 16
111111
[(google.api.field_behavior) = OUTPUT_ONLY];
112+
113+
// Output only. Reserved for future use.
114+
bool satisfies_pzs = 18 [(google.api.field_behavior) = OUTPUT_ONLY];
115+
116+
// Output only. Reserved for future use.
117+
bool satisfies_pzi = 19 [(google.api.field_behavior) = OUTPUT_ONLY];
112118
}
113119

114120
// Represents the spec of a CustomJob.
@@ -352,16 +358,16 @@ message PythonPackageSpec {
352358
// All parameters related to queuing and scheduling of custom jobs.
353359
message Scheduling {
354360
// Optional. This determines which type of scheduling strategy to use. Right
355-
// now users have two options such as ON_DEMAND which will use regular on
356-
// demand resources to schedule the job, the other is LOW_COST which would
361+
// now users have two options such as STANDARD which will use regular on
362+
// demand resources to schedule the job, the other is SPOT which would
357363
// leverage spot resources alongwith regular resources to schedule
358364
// the job.
359365
enum Strategy {
360-
// Strategy will default to ON_DEMAND.
366+
// Strategy will default to STANDARD.
361367
STRATEGY_UNSPECIFIED = 0;
362368

363369
// Regular on-demand provisioning strategy.
364-
ON_DEMAND = 1;
370+
ON_DEMAND = 1 [deprecated = true];
365371

366372
// Low cost by making potential use of spot resources.
367373
LOW_COST = 2 [deprecated = true];

‎google/cloud/aiplatform/v1beta1/entity_type.proto‎

Lines changed: 6 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -95,4 +95,10 @@ message EntityType {
9595
// values older than `offline_storage_ttl_days` since the feature generation
9696
// time. If unset (or explicitly set to 0), default to 4000 days TTL.
9797
int32 offline_storage_ttl_days = 10 [(google.api.field_behavior) = OPTIONAL];
98+
99+
// Output only. Reserved for future use.
100+
bool satisfies_pzs = 11 [(google.api.field_behavior) = OUTPUT_ONLY];
101+
102+
// Output only. Reserved for future use.
103+
bool satisfies_pzi = 12 [(google.api.field_behavior) = OUTPUT_ONLY];
98104
}

‎google/cloud/aiplatform/v1beta1/feature_group.proto‎

Lines changed: 15 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -40,6 +40,14 @@ message FeatureGroup {
4040

4141
// Input source type for BigQuery Tables and Views.
4242
message BigQuery {
43+
message TimeSeries {
44+
// Optional. Column hosting timestamp values for a time-series source.
45+
// Will be used to determine the latest `feature_values` for each entity.
46+
// Optional. If not provided, column named `feature_timestamp` of
47+
// type `TIMESTAMP` will be used.
48+
string timestamp_column = 1 [(google.api.field_behavior) = OPTIONAL];
49+
}
50+
4351
// Required. Immutable. The BigQuery source URI that points to either a
4452
// BigQuery Table or View.
4553
BigQuerySource big_query_source = 1 [
@@ -51,6 +59,13 @@ message FeatureGroup {
5159
// If not provided defaults to `entity_id`.
5260
repeated string entity_id_columns = 2
5361
[(google.api.field_behavior) = OPTIONAL];
62+
63+
// Optional. If the source is a time-series source, this can be set to
64+
// control how downstream sources (ex:
65+
// [FeatureView][google.cloud.aiplatform.v1beta1.FeatureView] ) will treat
66+
// time-series sources. If not set, will treat the source as a time-series
67+
// source with `feature_timestamp` as timestamp column and no scan boundary.
68+
TimeSeries time_series = 4 [(google.api.field_behavior) = OPTIONAL];
5469
}
5570

5671
oneof source {

‎google/cloud/aiplatform/v1beta1/feature_online_store.proto‎

Lines changed: 6 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -179,4 +179,10 @@ message FeatureOnlineStore {
179179
// Optional. Customer-managed encryption key spec for data storage. If set,
180180
// online store will be secured by this key.
181181
EncryptionSpec encryption_spec = 13 [(google.api.field_behavior) = OPTIONAL];
182+
183+
// Output only. Reserved for future use.
184+
bool satisfies_pzs = 15 [(google.api.field_behavior) = OUTPUT_ONLY];
185+
186+
// Output only. Reserved for future use.
187+
bool satisfies_pzi = 16 [(google.api.field_behavior) = OUTPUT_ONLY];
182188
}

‎google/cloud/aiplatform/v1beta1/feature_view.proto‎

Lines changed: 6 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -324,4 +324,10 @@ message FeatureView {
324324
// bigquery.dataViewer should be granted to this service account to allow
325325
// Vertex AI Feature Store to sync data to the online store.
326326
string service_account_email = 13 [(google.api.field_behavior) = OUTPUT_ONLY];
327+
328+
// Output only. Reserved for future use.
329+
bool satisfies_pzs = 19 [(google.api.field_behavior) = OUTPUT_ONLY];
330+
331+
// Output only. Reserved for future use.
332+
bool satisfies_pzi = 20 [(google.api.field_behavior) = OUTPUT_ONLY];
327333
}

‎google/cloud/aiplatform/v1beta1/feature_view_sync.proto‎

Lines changed: 6 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -67,4 +67,10 @@ message FeatureViewSync {
6767

6868
// Output only. Summary of the sync job.
6969
SyncSummary sync_summary = 6 [(google.api.field_behavior) = OUTPUT_ONLY];
70+
71+
// Output only. Reserved for future use.
72+
bool satisfies_pzs = 7 [(google.api.field_behavior) = OUTPUT_ONLY];
73+
74+
// Output only. Reserved for future use.
75+
bool satisfies_pzi = 8 [(google.api.field_behavior) = OUTPUT_ONLY];
7076
}

‎google/cloud/aiplatform/v1beta1/featurestore.proto‎

Lines changed: 6 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -149,4 +149,10 @@ message Featurestore {
149149
// Optional. Customer-managed encryption key spec for data storage. If set,
150150
// both of the online and offline data storage will be secured by this key.
151151
EncryptionSpec encryption_spec = 10 [(google.api.field_behavior) = OPTIONAL];
152+
153+
// Output only. Reserved for future use.
154+
bool satisfies_pzs = 14 [(google.api.field_behavior) = OUTPUT_ONLY];
155+
156+
// Output only. Reserved for future use.
157+
bool satisfies_pzi = 15 [(google.api.field_behavior) = OUTPUT_ONLY];
152158
}

‎google/cloud/aiplatform/v1beta1/hyperparameter_tuning_job.proto‎

Lines changed: 6 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -113,4 +113,10 @@ message HyperparameterTuningJob {
113113
// If this is set, then all resources created by the HyperparameterTuningJob
114114
// will be encrypted with the provided encryption key.
115115
EncryptionSpec encryption_spec = 17;
116+
117+
// Output only. Reserved for future use.
118+
bool satisfies_pzs = 19 [(google.api.field_behavior) = OUTPUT_ONLY];
119+
120+
// Output only. Reserved for future use.
121+
bool satisfies_pzi = 20 [(google.api.field_behavior) = OUTPUT_ONLY];
116122
}

‎google/cloud/aiplatform/v1beta1/index.proto‎

Lines changed: 6 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -125,6 +125,12 @@ message Index {
125125
// Immutable. Customer-managed encryption key spec for an Index. If set, this
126126
// Index and all sub-resources of this Index will be secured by this key.
127127
EncryptionSpec encryption_spec = 17 [(google.api.field_behavior) = IMMUTABLE];
128+
129+
// Output only. Reserved for future use.
130+
bool satisfies_pzs = 18 [(google.api.field_behavior) = OUTPUT_ONLY];
131+
132+
// Output only. Reserved for future use.
133+
bool satisfies_pzi = 19 [(google.api.field_behavior) = OUTPUT_ONLY];
128134
}
129135

130136
// A datapoint of Index.

0 commit comments

Comments
 (0)