build(deps): bump google-cloud-aiplatform from 1.143.0 to 1.145.0 in the google-cloud group across 1 directory

[dependabot]

Bumps the google-cloud group with 1 update in the / directory: google-cloud-aiplatform.

Updates google-cloud-aiplatform from 1.143.0 to 1.145.0

Release notes

Sourced from google-cloud-aiplatform's releases.

v1.145.0

1.145.0 (2026-04-01)

Features

• Add session_id to Create Session to allow custom session id (bea67c2)
• Eval SDK: Migrate model call method by genai SDK usage (ff5e246)
• Eval SDK: Migrate model call method by genai SDK usage in preview foler (ad36123)
• Limit metric registry support to only custom code execution metric and llm based metric in SDK (c12aedc)

Bug Fixes

• Increase default timeout to 600 seconds for ask_contexts and async_retrieve_contexts in VertexRagServiceClient. (3de2c1e)

v1.144.0

1.144.0 (2026-03-31)

⚠ BREAKING CHANGES

• Breaking change in experimental methods: create_from_bigquery and update_multimodal_dataset no longer automatically prepend a missing bq:// prefix for BigQuery URIs. When using the new function MultimodalDataset.set_bigquery_uri the prefix will still be added if needed.

Features

• Add metadata_filter usage to all retrieval and generator methods in rag_retrieval.py (841c597)
• Add container_spec to Reasoning Engine public protos (9a0eefb)
• Add container_spec to Reasoning Engine public protos (9a0eefb)
• Add support for container_spec in AgentEngines (da663c0)
• GenAI SDK client(multimodal) - Add metadata helpers to MultimodalDataset. (e164b19)
• Publish client batch config schema (9a0eefb)
• Refactor evaluation instance building and update LLM metric handler (7a3b436)
• Update the interface for custom code execution metric while maintaining remote_custom_function support for backward compatibility (f7733ec)

Bug Fixes

• Auto-created GCS staging bucket names are less predictable (1a33ad9)
• GenAI SDK client(multimodal) - Replace blocking calls in create_from_pandas with async versions. (2767273)
• Introduce timeout for outgoing A2A requests from AgentEngine (78525d2)
• Relax the requirement to specify class_methods if image_spec/container_spec is provided (6f7b12c)
• Standardize on the app_name in AdkApp (ee9fbe1)

Documentation

• Update API common type documentation (9a0eefb)

... (truncated)

Changelog

Sourced from google-cloud-aiplatform's changelog.

1.145.0 (2026-04-01)

Features

• Add session_id to Create Session to allow custom session id (bea67c2)
• Eval SDK: Migrate model call method by genai SDK usage (ff5e246)
• Eval SDK: Migrate model call method by genai SDK usage in preview foler (ad36123)
• Limit metric registry support to only custom code execution metric and llm based metric in SDK (c12aedc)

Bug Fixes

• Increase default timeout to 600 seconds for ask_contexts and async_retrieve_contexts in VertexRagServiceClient. (3de2c1e)

1.144.0 (2026-03-31)

⚠ BREAKING CHANGES

• create_from_bigquery and update_multimodal_dataset no longer automatically prepend a missing bq:// prefix for BigQuery URIs. When using the new function MultimodalDataset.set_bigquery_uri the prefix will still be added if needed.

Features

• Add metadata_filter usage to all retrieval and generator methods in rag_retrieval.py (841c597)
• Add container_spec to Reasoning Engine public protos (9a0eefb)
• Add container_spec to Reasoning Engine public protos (9a0eefb)
• Add support for container_spec in AgentEngines (da663c0)
• GenAI SDK client(multimodal) - Add metadata helpers to MultimodalDataset. (e164b19)
• Publish client batch config schema (9a0eefb)
• Refactor evaluation instance building and update LLM metric handler (7a3b436)
• Update the interface for custom code execution metric while maintaining remote_custom_function support for backward compatibility (f7733ec)

Bug Fixes

• Auto-created GCS staging bucket names are less predictable (1a33ad9)
• GenAI SDK client(multimodal) - Replace blocking calls in create_from_pandas with async versions. (2767273)
• Introduce timeout for outgoing A2A requests from AgentEngine (78525d2)
• Relax the requirement to specify class_methods if image_spec/container_spec is provided (6f7b12c)
• Standardize on the app_name in AdkApp (ee9fbe1)

Documentation

• Update API common type documentation (9a0eefb)

Miscellaneous Chores

... (truncated)

Commits

• 7a8f703 chore(main): release 1.145.0 (#6524)
• ad36123 feat: Eval SDK: Migrate model call method by genai SDK usage in preview foler
• 3de2c1e fix: Increase default timeout to 600 seconds for ask_contexts and async_retri...
• ff5e246 feat: Eval SDK: Migrate model call method by genai SDK usage
• bea67c2 feat: Add session_id to Create Session to allow custom session id
• c12aedc feat: Limit metric registry support to only custom code execution metric and ...
• 1fba45b chore(main): release 1.144.0 (#6486)
• 1a33ad9 fix: auto-created GCS staging bucket names are less predictable
• f7733ec feat: Update the interface for custom code execution metric while maintaining...
• 2ad586d chore: release 1.144.0
• Additional commits viewable in compare view

[github-actions]

All CI checks passed for this core dependency update.

@em-pe Please review and merge manually.

google-cloud-aiplatform 1.143.0 → 1.145.0

Breaking changes:

• create_from_bigquery and update_multimodal_dataset (experimental) no longer automatically prepend bq:// prefix for BigQuery URIs — not relevant to kedro-vertexai

Deprecations:

• None

Notable changes:

• Bug fix: Auto-created GCS staging bucket names are now less predictable (v1.144.0) — low impact for kedro-vertexai, but worth noting if your pipelines rely on predictable staging bucket naming
• Bug fix: Increased default timeout to 600s for ask_contexts/async_retrieve_contexts in VertexRagServiceClient (v1.145.0)
• Feature: container_spec added to Reasoning Engine / AgentEngine public protos (v1.144.0)
• Feature: session_id support added to Create Session API (v1.145.0)
• Other changes are in Eval SDK, RAG retrieval, and GenAI multimodal dataset APIs — not used by kedro-vertexai

Risk assessment: Low — all changes are in areas unrelated to kedro-vertexai's core functionality (pipeline execution, KFP compilation, GCP Pipelines deployment). The one breaking change only affects experimental multimodal dataset methods not used by this plugin.

Full changelog

1.145.0 (2026-04-01)

Features:

• Add session_id to Create Session to allow custom session id
• Eval SDK: Migrate model call method by genai SDK usage
• Eval SDK: Migrate model call method by genai SDK usage in preview folder
• Limit metric registry support to only custom code execution metric and llm based metric in SDK

Bug Fixes:

• Increase default timeout to 600 seconds for ask_contexts and async_retrieve_contexts in VertexRagServiceClient

---

1.144.0 (2026-03-31)

⚠ BREAKING CHANGES:

• create_from_bigquery and update_multimodal_dataset no longer automatically prepend a missing bq:// prefix for BigQuery URIs. When using the new function MultimodalDataset.set_bigquery_uri the prefix will still be added if needed.

Features:

• Add metadata_filter usage to all retrieval and generator methods in rag_retrieval.py
• Add container_spec to Reasoning Engine public protos
• Add support for container_spec in AgentEngines
• GenAI SDK client(multimodal) - Add metadata helpers to MultimodalDataset
• Publish client batch config schema
• Refactor evaluation instance building and update LLM metric handler
• Update the interface for custom code execution metric while maintaining remote_custom_function support for backward compatibility

Bug Fixes:

• Auto-created GCS staging bucket names are less predictable
• GenAI SDK client(multimodal) - Replace blocking calls in create_from_pandas with async versions
• Introduce timeout for outgoing A2A requests from AgentEngine
• Relax the requirement to specify class_methods if image_spec/container_spec is provided
• Standardize on the app_name in AdkApp

Note

🔒 Integrity filter blocked 2 items

The following items were blocked because they don't meet the GitHub integrity level.

• build(deps): bump google-cloud-aiplatform from 1.143.0 to 1.145.0 in the google-cloud group across 1 directory #310 pull_request_read: has lower integrity than agent requires. Agent would need to drop integrity tags [unapproved:all approved:all] to trust this resource.
• #240 list_pull_requests: has lower integrity than agent requires. Agent would need to drop integrity tags [approved:all] to trust this resource.

To allow these resources, lower min-integrity in your GitHub frontmatter:

tools:
  github:
    min-integrity: approved  # merged | approved | unapproved | none

Generated by Core Dependency Review Agent · ◷

[github-actions]

All CI checks passed for this core dependency update.

@em-pe Please review and merge manually.

google-cloud-aiplatform 1.143.0 → 1.145.0

Breaking changes:

• create_from_bigquery and update_multimodal_dataset (experimental) no longer auto-prepend bq:// prefix for BigQuery URIs — not relevant to kedro-vertexai (pipeline execution API is unaffected)

Deprecations:

• None

Notable changes:

v1.145.0 (2026-04-01):

• Add session_id parameter to Create Session API for custom session IDs
• Eval SDK: migrate model call methods to genai SDK
• Limit metric registry to custom code execution and LLM-based metrics
• Fix: increase default timeout to 600s for ask_contexts/async_retrieve_contexts in VertexRagServiceClient

v1.144.0 (2026-03-31):

• Fix: auto-created GCS staging bucket names are now less predictable (minor consideration if code relies on deterministic bucket names)
• Add container_spec support to Reasoning Engine / AgentEngines
• Add metadata_filter to RAG retrieval methods
• GenAI SDK multimodal dataset helpers and async fixes
• AgentEngine: timeout for outgoing A2A requests

Risk assessment: Low — all changes are in GenAI/Eval/RAG/AgentEngine subcomponents; none touch the Vertex AI Pipelines API that kedro-vertexai uses for pipeline compilation and execution.

Full changelog

1.145.0 (2026-04-01)

Features

• Add session_id to Create Session to allow custom session id (bea67c2)
• Eval SDK: Migrate model call method by genai SDK usage (ff5e246)
• Eval SDK: Migrate model call method by genai SDK usage in preview folder (ad36123)
• Limit metric registry support to only custom code execution metric and llm based metric in SDK (c12aedc)

Bug Fixes

• Increase default timeout to 600 seconds for ask_contexts and async_retrieve_contexts in VertexRagServiceClient (3de2c1e)

---

1.144.0 (2026-03-31)

⚠ BREAKING CHANGES

• Breaking change in experimental methods: create_from_bigquery and update_multimodal_dataset no longer automatically prepend a missing bq:// prefix for BigQuery URIs. When using the new function MultimodalDataset.set_bigquery_uri the prefix will still be added if needed.

Features

• Add metadata_filter usage to all retrieval and generator methods in rag_retrieval.py (841c597)
• Add container_spec to Reasoning Engine public protos (9a0eefb)
• Add support for container_spec in AgentEngines (da663c0)
• GenAI SDK client(multimodal) - Add metadata helpers to MultimodalDataset (e164b19)
• Publish client batch config schema (9a0eefb)
• Refactor evaluation instance building and update LLM metric handler (7a3b436)
• Update the interface for custom code execution metric while maintaining remote_custom_function support for backward compatibility (f7733ec)

Bug Fixes

• Auto-created GCS staging bucket names are less predictable (1a33ad9)
• GenAI SDK client(multimodal) - Replace blocking calls in create_from_pandas with async versions (2767273)
• Introduce timeout for outgoing A2A requests from AgentEngine (78525d2)
• Relax the requirement to specify class_methods if image_spec/container_spec is provided (6f7b12c)
• Standardize on the app_name in AdkApp (ee9fbe1)

Note

🔒 Integrity filter blocked 3 items

The following items were blocked because they don't meet the GitHub integrity level.

• build(deps): bump google-cloud-aiplatform from 1.143.0 to 1.145.0 in the google-cloud group across 1 directory #310 pull_request_read: has lower integrity than agent requires. Agent would need to drop integrity tags [approved:all unapproved:all] to trust this resource.
• #240 list_pull_requests: has lower integrity than agent requires. Agent would need to drop integrity tags [approved:all] to trust this resource.
• build(deps): bump google-cloud-aiplatform from 1.143.0 to 1.145.0 in the google-cloud group across 1 directory #310 pull_request_read: has lower integrity than agent requires. Agent would need to drop integrity tags [unapproved:all approved:all] to trust this resource.

To allow these resources, lower min-integrity in your GitHub frontmatter:

tools:
  github:
    min-integrity: approved  # merged | approved | unapproved | none

Generated by Core Dependency Review Agent · ◷

[dependabot]

Looks like google-cloud-aiplatform is updatable in another way, so this is no longer needed.