Releases: langchain-ai/langchain-google
Releases · langchain-ai/langchain-google
libs/community/v2.0.0.dev2
What's Changed
- fix BQ VectorStore dataset permission and table schema info by @eliasecchig in #464
- vertexai: added support for custom ids and sprinkled in some docstring updates by @rubenschulze in #467
- Added support for Google Cloud Document AI Layout Parser by @RajeshThallam in #294
- genai: add mkdir -p option for make lint by @nobu007 in #470
- genai, vertex, community[patch]: backward-compatible pydantic v2 changes by @ccurme in #472
- remove integration tests for bison by @lkuligin in #477
- genai: use SecretStr for api_key by @nobu007 in #397
- community: Add complex filters bigqueryvectorstore (SQL filters) by @Freezaa9 in #448
- added reduced dimensionanility for multimodal embeddings by @lkuligin in #476
- genai[patch]: add standard tests by @ccurme in #478
- Revert "genai[patch]: add standard tests" by @ccurme in #479
- genai[patch]: fix standard tests by @ccurme in #480
- infra[patch]: support dev + rc releases by @ccurme in #483
- infra[patch]: add integration test workflow by @ccurme in #484
New Contributors
- @rubenschulze made their first contribution in #467
- @RajeshThallam made their first contribution in #294
Full Changelog: libs/vertexai/v1.0.10...libs/community/v2.0.0.dev2
libs/vertexai/v1.0.10
What's Changed
Full Changelog: libs/community/v1.0.8...libs/vertexai/v1.0.10
libs/vertexai/v1.0.9
What's Changed
- genai and vertexai: implement ls_params for LLMs by @ccurme in #446
- added support for mistral maas models by @lkuligin in #447
- fix default params + added support for Gemma 2B by @lkuligin in #451
- genai: Add support for allOf by @gramliu in #453
- vertex: fix SafetySetting import by @kardiff18 in #450
- fix condition for gemini advanced model family by @lkuligin in #454
- Tests cleanup by @lkuligin in #455
- test for mistral tools by @lkuligin in #456
- Llama maas by @lkuligin in #458
- release by @lkuligin in #461
New Contributors
Full Changelog: libs/genai/v1.0.9...libs/vertexai/v1.0.9
libs/genai/v1.0.10
What's Changed
- genai and vertexai: implement ls_params for LLMs by @ccurme in #446
- added support for mistral maas models by @lkuligin in #447
- fix default params + added support for Gemma 2B by @lkuligin in #451
- genai: Add support for allOf by @gramliu in #453
- vertex: fix SafetySetting import by @kardiff18 in #450
- fix condition for gemini advanced model family by @lkuligin in #454
- Tests cleanup by @lkuligin in #455
- test for mistral tools by @lkuligin in #456
- Llama maas by @lkuligin in #458
- release by @lkuligin in #461
New Contributors
Full Changelog: libs/genai/v1.0.9...libs/genai/v1.0.10
libs/community/v1.0.8
What's Changed
- genai and vertexai: implement ls_params for LLMs by @ccurme in #446
- added support for mistral maas models by @lkuligin in #447
- fix default params + added support for Gemma 2B by @lkuligin in #451
- genai: Add support for allOf by @gramliu in #453
- vertex: fix SafetySetting import by @kardiff18 in #450
- fix condition for gemini advanced model family by @lkuligin in #454
- Tests cleanup by @lkuligin in #455
- test for mistral tools by @lkuligin in #456
- Llama maas by @lkuligin in #458
- release by @lkuligin in #461
New Contributors
Full Changelog: libs/genai/v1.0.9...libs/community/v1.0.8
libs/genai/v1.0.9
What's Changed
- genai[patch]: with_structured_output support pydantic v2 by @baskaryan in #414
- add rate limiting to integration tests by @eyurtsev in #408
- genai: fix Makefile about ruff check by @nobu007 in #419
- GenerateContentRequest.tools.function_declarations.parameters set None when function parameters is empty by @bimslab in #416
- vertex[patch]: fix streaming for gemini-1.5-pro + vision by @ccurme in #418
- Update gemma.py by @Tungway1990 in #412
- Fix vertexaimodel by @pprados in #406
- Update drive.py to only use drive.file scope by @walterheck in #424
- docs: add PULL_REQUEST_TEMPLATE.md by @nobu007 in #398
- fix: update table_schema when instantiating already existing BigQueryVectorStore by @Freezaa9 in #429
- added support for allOf by @lkuligin in #431
- fix: Standardize default parameters across Gemini and Gemini Advanced models by @mitraan-deshaw in #428
- added deprecation warning for a default model_name by @lkuligin in #433
- Add Multimodal fc support by @kardiff18 in #404
- genai[patch]: support media inputs by @baskaryan in #442
- make ChatGoogleGenerativeAI serializable by @lkuligin in #443
- genai[patch]: Release 1.0.9 by @baskaryan in #445
New Contributors
- @eyurtsev made their first contribution in #408
- @Tungway1990 made their first contribution in #412
- @pprados made their first contribution in #406
- @walterheck made their first contribution in #424
- @Freezaa9 made their first contribution in #429
- @mitraan-deshaw made their first contribution in #428
Full Changelog: libs/vertexai/v1.0.8...libs/genai/v1.0.9
libs/vertexai/v1.0.8
What's Changed
- Update the FS logic to manage null values parsing by @ThibaultScalbert in #391
- Support chaining in vision models by @jzaldi in #393
- Multithreaded GCSDocumentStorage uploads and downloads by @atognolag in #318
- add finish_reason to gemini gen info by @michaelchia in #380
- ChatVertexAI support async streaming for gemini model by @bimslab in #400
- Fix support for additional headers by @tomaszZdunek in #396
- vertexai[patch]: bind tools in oai format by @baskaryan in #405
- vertexai[patch]: Release 1.0.8 by @baskaryan in #413
New Contributors
- @ThibaultScalbert made their first contribution in #391
- @atognolag made their first contribution in #318
- @michaelchia made their first contribution in #380
- @bimslab made their first contribution in #400
Full Changelog: libs/vertexai/v1.0.7...libs/vertexai/v1.0.8
libs/vertexai/v1.0.7
What's Changed
Full Changelog: libs/community/v1.0.7...libs/vertexai/v1.0.7
libs/genai/v1.0.8
What's Changed
- fix: ChatAnthropicVertex returning None during structured output by @kiarina in #332
- fixed sequence of multiple AI messages by @alx13 in #335
- fix embeddings with max_batch_size by @lkuligin in #338
- chore: add stop_reason in ChatAnthropicVertex stream response by @kiarina in #342
- added boosting for vertex search by @lkuligin in #346
- Add client opts for Vertex Rank and Vertex Check Grounding by @Abhishekbhagwat in #347
- Add enforce json schema to ChatVertexAI by @jzaldi in #348
- genai[patch]: fix tool calling, add docstrings by @baskaryan in #349
- [Breaking] change the behavior of ChatVertexAI.with_structured_output when specifying a dict by @kiarina in #333
- [ChatVertexAI] Get
api_transport
from initializer when not provided by @jzaldi in #350 - release permissions by @efriis in #359
- turn off tests for medlm-latest by @lkuligin in #362
- Add InternalServerError to retry_errors in VertexAIEmbeddings by @TupleType in #358
- vertexai[patch], genai[patch]: use tool name explicitly in with struc… by @baskaryan in #341
- Added new gemini family model MedLM-Large-1.5-preview by @smannathan in #336
- fixed integration tests by @lkuligin in #373
- Add domain-wide service authentification and authorization by @mamisoa in #370
- Fix tests by @gmogr in #360
- fix async calls without event_loop by @lkuligin in #375
- Fix optional fields in Pydantic models by @orkhank in #363
- feat: add google credential support to ChatAnthropicVertex by @kumaray-deshaw in #377
- Fix: BigQueryVectorStore crashes on BQ tables with more than 5000 rows by @thomasfitze in #365
- Add request_metadata to vertexai.init by @tomaszZdunek in #378
- Fix bug caused by downgrading pydantic from 2 to 1 by @lspataroG in #374
- Add Context catching to VertexAI Models by @jzaldi in #352
- vertexai[patch], genai[patch]: specify ToolCall "type" by @baskaryan in #366
- genai[patch]: update vision model used in integration tests by @ccurme in #385
- Support page number for segments by @julienrzeznik in #384
- fix: wrong page number in document metadata by @ShuJ-i-e in #381
- fix calling multiple tools + refactor to make closer to VertexAI impl… by @lkuligin in #387
- fixed test by @lkuligin in #390
- release by @lkuligin in #388
New Contributors
- @kiarina made their first contribution in #332
- @TupleType made their first contribution in #358
- @mamisoa made their first contribution in #370
- @orkhank made their first contribution in #363
- @kumaray-deshaw made their first contribution in #377
- @thomasfitze made their first contribution in #365
- @tomaszZdunek made their first contribution in #378
- @julienrzeznik made their first contribution in #384
- @ShuJ-i-e made their first contribution in #381
Full Changelog: libs/community/v1.0.6...libs/genai/v1.0.8
libs/community/v1.0.7
What's Changed
- fix: ChatAnthropicVertex returning None during structured output by @kiarina in #332
- fixed sequence of multiple AI messages by @alx13 in #335
- fix embeddings with max_batch_size by @lkuligin in #338
- chore: add stop_reason in ChatAnthropicVertex stream response by @kiarina in #342
- added boosting for vertex search by @lkuligin in #346
- Add client opts for Vertex Rank and Vertex Check Grounding by @Abhishekbhagwat in #347
- Add enforce json schema to ChatVertexAI by @jzaldi in #348
- genai[patch]: fix tool calling, add docstrings by @baskaryan in #349
- [Breaking] change the behavior of ChatVertexAI.with_structured_output when specifying a dict by @kiarina in #333
- [ChatVertexAI] Get
api_transport
from initializer when not provided by @jzaldi in #350 - release permissions by @efriis in #359
- turn off tests for medlm-latest by @lkuligin in #362
- Add InternalServerError to retry_errors in VertexAIEmbeddings by @TupleType in #358
- vertexai[patch], genai[patch]: use tool name explicitly in with struc… by @baskaryan in #341
- Added new gemini family model MedLM-Large-1.5-preview by @smannathan in #336
- fixed integration tests by @lkuligin in #373
- Add domain-wide service authentification and authorization by @mamisoa in #370
- Fix tests by @gmogr in #360
- fix async calls without event_loop by @lkuligin in #375
- Fix optional fields in Pydantic models by @orkhank in #363
- feat: add google credential support to ChatAnthropicVertex by @kumaray-deshaw in #377
- Fix: BigQueryVectorStore crashes on BQ tables with more than 5000 rows by @thomasfitze in #365
- Add request_metadata to vertexai.init by @tomaszZdunek in #378
- Fix bug caused by downgrading pydantic from 2 to 1 by @lspataroG in #374
- Add Context catching to VertexAI Models by @jzaldi in #352
- vertexai[patch], genai[patch]: specify ToolCall "type" by @baskaryan in #366
- genai[patch]: update vision model used in integration tests by @ccurme in #385
- Support page number for segments by @julienrzeznik in #384
- fix: wrong page number in document metadata by @ShuJ-i-e in #381
- fix calling multiple tools + refactor to make closer to VertexAI impl… by @lkuligin in #387
- fixed test by @lkuligin in #390
- release by @lkuligin in #388
New Contributors
- @kiarina made their first contribution in #332
- @TupleType made their first contribution in #358
- @mamisoa made their first contribution in #370
- @orkhank made their first contribution in #363
- @kumaray-deshaw made their first contribution in #377
- @thomasfitze made their first contribution in #365
- @tomaszZdunek made their first contribution in #378
- @julienrzeznik made their first contribution in #384
- @ShuJ-i-e made their first contribution in #381
Full Changelog: libs/community/v1.0.6...libs/community/v1.0.7