2166 Commits (193f107cb5b01ec19ec5c56bae2c426c54791808)

Author	SHA1	Message	Date
Harrison Chase	193f107cb5	add methods to deserialize prompts that were old (#14857)	9 months ago
Bagatur	714bef0cb6	langchain[patch]: Release 0.0.351 (#14867)	9 months ago
Bagatur	61ad0e8be9	community[patch]: Release 0.0.4 (#14864)	9 months ago
Bob Lin	5de1dc72b9	community[patch]: Update Tongyi default model_name (#14844) ... <img width="1305" alt="Screenshot 2023-12-18 at 9 54 01 PM" src="https://github.com/langchain-ai/langchain/assets/10000925/c943fd81-cd48-46eb-8dff-4680424d9ba9"> The current model is no longer available.	9 months ago
William FH	5fc2c578cf	[Bugfix] Ensure tool output is a str, for OAI Assistant (#14830) ... Tool outputs have to be strings apparently. Ensure they are formatted correctly before passing as intermediate steps. ``` BadRequestError: Error code: 400 - {'error': {'message': '1 validation error for Request\nbody -> tool_outputs -> 0 -> output\n str type expected (type=type_error.str)', 'type': 'invalid_request_error', 'param': None, 'code': None}} ```	9 months ago
William FH	bbc98a234d	Update parser (#14831) ... Gpt-3.5 sometimes calls with empty string arguments instead of `{}` I'd assume it's because the typescript representation on their backend makes it a bit ambiguous.	9 months ago
Vlad Kolesnikov	11fda490ca	community[minor]: New model parameters and dynamic batching for VertexAIEmbeddings (#13999) ... - **Description:** VertexAIEmbeddings performance improvements - **Twitter handle:** @vladkol ## Improvements - Dynamic batch size, starting from 250, lowering down to 5. Batch size varies across regions. Some regions support larger batches, and it significantly improves performance. When running large batches of texts in `us-central1`, performance gain can be up to 3.5x. The dynamic batching also makes sure every batch is below 20K token limit. - New model parameter `embeddings_type` that translates to `task_type` parameter of the API. Newer model versions support [different embeddings task types](https://cloud.google.com/vertex-ai/docs/generative-ai/embeddings/get-text-embeddings#api_changes_to_models_released_on_or_after_august_2023).	9 months ago
William FH	2d91d2b978	community: Add logprobs in gen output (#14826) ... Now that it's supported again for OAI chat models . Shame this wouldn't include it in the `.invoke()` output though (it's not included in the message itself). Would need to do a follow-up for that to be the case	9 months ago
Dmitry Tyumentsev	78ae276df7	community[patch]: fix agenerate return value (#14815) ... Fixed: - `_agenerate` return value in the YandexGPT Chat Model - duplicate line in the documentation Co-authored-by: Dmitry Tyumentsev <dmitry.tyumentsev@raftds.com>	9 months ago
sujeet	f1d3f29bc4	community[patch]: support for Sybase SQL anywhere added. (#14821) ... - **Description:** support for Sybase SQL anywhere added in sql_database.py file at path langchain\libs\community\langchain_community\utilities - **Issue:** It will resolve default schema setting for Sybase SQL anywhere - **Dependencies:** No, - **Tag maintainer:** @baskaryan, @eyurtsev, @hwchase17, - **Twitter handle:** NA --------- Co-authored-by: learn360sujeet <121271779+learn360sujeet@users.noreply.github.com> Co-authored-by: Bagatur <baskaryan@gmail.com>	9 months ago
Erick Friis	8a07c56313	docs: developer docs (#14776) ... Builds out a developer documentation section in the docs - Links it from contributing.md - Adds an initial guide on how to contribute an integration --------- Co-authored-by: Bagatur <baskaryan@gmail.com>	9 months ago
William FH	01693b291e	Permit updates in indexing (#14482)	9 months ago
Noah Stapp	34e6f3ff72	community[patch]: Implement similarity_score_threshold for MongoDB Vector Store (#14740) ... Adds the option for `similarity_score_threshold` when using `MongoDBAtlasVectorSearch` as a vector store retriever. Example use: ``` vector_search = MongoDBAtlasVectorSearch.from_documents(...) qa_retriever = vector_search.as_retriever( search_type="similarity_score_threshold", search_kwargs={ "score_threshold": 0.5, } ) qa = RetrievalQA.from_chain_type( llm=OpenAI(), chain_type="stuff", retriever=qa_retriever, ) docs = qa({"query": "..."}) ``` I've tested this feature locally, using a MongoDB Atlas Cluster with a vector search index.	9 months ago
Dmitry Tyumentsev	dcead816df	community[patch]: Update YandexGPT API (#14773) ... Update LLMand Chat model to use new api version --------- Co-authored-by: Dmitry Tyumentsev <dmitry.tyumentsev@raftds.com>	9 months ago
Lance Martin	42421860bc	Add image support for Ollama (#14713) ... Support [LLaVA](https://ollama.ai/library/llava): * Upgrade Ollama * `ollama pull llava` Ensure compatibility with [image prompt template](https://github.com/langchain-ai/langchain/pull/14263) --------- Co-authored-by: jacoblee93 <jacoblee93@gmail.com>	9 months ago
Harrison Chase	16399fd61d	langchain[patch]: remove unused imports (#14680) ... Co-authored-by: Bagatur <baskaryan@gmail.com>	9 months ago
Karim Lalani	a0064330b1	community[minor]: Add SurrealDB vectorstore (#13331) ... **Description:** Vectorstore implementation around [SurrealDB](https://www.surrealdb.com) --------- Co-authored-by: Bagatur <baskaryan@gmail.com>	9 months ago
William FH	c5296fd42c	[Documentation] Updates to NVIDIA Playground/Foundation Model naming.… (#14770) ... … (#14723) - **Description:** Minor updates per marketing requests. Namely, name decisions (AI Foundation Models / AI Playground) - **Tag maintainer:** @hinthornw Do want to pass around the PR for a bit and ask a few more marketing questions before merge, but just want to make sure I'm not working in a vacuum. No major changes to code functionality intended; the PR should be for documentation and only minor tweaks. Note: QA model is a bit borked across staging/prod right now. Relevant teams have been informed and are looking into it, and I'm placeholdered the response to that of a working version in the notebook. Co-authored-by: Vadim Kudlay <32310964+VKudlay@users.noreply.github.com>	9 months ago
William FH	4855964332	Fix OAI Tool Message (#14746) ... See format here: https://platform.openai.com/docs/guides/function-calling/parallel-function-calling It expects a "name" argument, which we aren't providing by default.  Alternative is to add the 'name' field directly to the message if people prefer.	9 months ago
William FH	e3132a7efc	[Evals] End project (#14324) ... Also does some cleanup. Now that we support updating/ending projects, do this automatically. Then you can edit the name of the project in the app.	9 months ago
William FH	93c7eb4e6b	[Tracing] String Stacktrace (#14131) ... Add full stacktrace	9 months ago
Leonid Kuligin	7f42811e14	google-genai[patch], community[patch]: Added support for new Google GenerativeAI models (#14530) ... Replace this entire comment with: - **Description:** added support for new Google GenerativeAI models - **Twitter handle:** lkuligin --------- Co-authored-by: Erick Friis <erick@langchain.dev>	9 months ago
Bagatur	b802dd96f2	core[patch]: Release 0.1.1 (#14738)	9 months ago
William FH	9d4100f915	Revert "[Hub|tracing] Tag hub prompts" (#14735) ... Reverts langchain-ai/langchain#14720	9 months ago
Erick Friis	9fb26a2a71	community[patch]: fix pgvector sqlalchemy (#14726) ... Fixes #14699	9 months ago
Bagatur	1cec0afc62	google-genai[patch]: add google-genai integration deps and extras (#14731)	9 months ago
William FH	852b9ca494	[Hub|tracing] Tag hub prompts (#14720) ... If you're using the hub, you'll likely be interested in tracking the commit/object when tracing. This PR adds it to the config	9 months ago
William FH	451c5d1d8c	[Integration] NVIDIA AI Playground (#14648) ... Description: Added NVIDIA AI Playground Initial support for a selection of models (Llama models, Mistral, etc.) Dependencies: These models do depend on the AI Playground services in NVIDIA NGC. API keys with a significant amount of trial compute are available (10K queries as of the time of writing). H/t to @VKudlay	9 months ago
William FH	1e21a3f7ed	[Partner] Gemini Embeddings (#14690) ... Add support for Gemini embeddings in the langchain-google-genai package	9 months ago
Funkeke	ea99612caa	community[patch]: fix dashvector endpoint params error (#14484) ... <!-- Thank you for contributing to LangChain! Replace this entire comment with: - **Description:** a description of the change, - **Issue:** the issue # it fixes (if applicable), - **Dependencies:** any dependencies required for this change, - **Tag maintainer:** for a quicker response, tag the relevant maintainer (see below), - **Twitter handle:** we announce bigger features on Twitter. If your PR gets announced, and you'd like a mention, we'll gladly shout you out! Please make sure your PR is passing linting and testing before submitting. Run `make format`, `make lint` and `make test` to check this locally. See contribution guidelines for more information on how to write/run tests, lint, etc: https://github.com/langchain-ai/langchain/blob/master/.github/CONTRIBUTING.md If you're adding a new integration, please include: 1. a test for the integration, preferably unit tests that do not rely on network access, 2. an example notebook showing its use. It lives in `docs/extras` directory. If no one reviews your PR within a few days, please @-mention one of @baskaryan, @eyurtsev, @hwchase17. --> Co-authored-by: fangkeke <3339698829@qq.com> Co-authored-by: Harrison Chase <hw.chase.17@gmail.com>	9 months ago
Bob Lin	dce3c74905	community[patch]: Correct type annotation for azure_ad_token_provider Closed: #14402 (#14432) ... Description Fix https://github.com/langchain-ai/langchain/issues/14402, Similar changes: https://github.com/langchain-ai/langchain/pull/14166 Twitter handle [lin_bob57617](https://twitter.com/lin_bob57617)	9 months ago
Fran Cirka	8a4162d15e	community[patch]: Fixed issue with importing Row from sqlalchemy (#14488) ... - **Description:** Fixed import of Row in cache.py, - **Issue:** the issue # #13464 https://creditone.us.to/langchain-ai/langchain/issues/13464, - **Dependencies:** None, - **Twitter handle:** @frankybridman Co-authored-by: Harrison Chase <hw.chase.17@gmail.com>	9 months ago
Bagatur	47451951a1	core[patch]: Fix runnable with message history (#14629) ... Fix bug shown in #14458. Namely, that saving inputs to history fails when the input to base runnable is a list of messages	9 months ago
Bagatur	73382a579f	google-genai[patch]: Release 0.0.2 (#14677)	9 months ago
Nuno Campos	a16f4a318f	\Fix tool_calls message merge (#14613) ... <!-- Thank you for contributing to LangChain! Replace this entire comment with: - **Description:** a description of the change, - **Issue:** the issue # it fixes (if applicable), - **Dependencies:** any dependencies required for this change, - **Tag maintainer:** for a quicker response, tag the relevant maintainer (see below), - **Twitter handle:** we announce bigger features on Twitter. If your PR gets announced, and you'd like a mention, we'll gladly shout you out! Please make sure your PR is passing linting and testing before submitting. Run `make format`, `make lint` and `make test` to check this locally. See contribution guidelines for more information on how to write/run tests, lint, etc: https://github.com/langchain-ai/langchain/blob/master/.github/CONTRIBUTING.md If you're adding a new integration, please include: 1. a test for the integration, preferably unit tests that do not rely on network access, 2. an example notebook showing its use. It lives in `docs/extras` directory. If no one reviews your PR within a few days, please @-mention one of @baskaryan, @eyurtsev, @hwchase17. -->	9 months ago
William FH	405d111da6	[Partner] Add langchain-google-genai package (gemini) (#14621) ... Add a new ChatGoogleGenerativeAI class in a `langchain-google-genai` package. Still todo: add a deprecation warning in PALM --------- Co-authored-by: Erick Friis <erick@langchain.dev> Co-authored-by: Leonid Kuligin <lkuligin@yandex.ru> Co-authored-by: Bagatur <baskaryan@gmail.com>	9 months ago
Bagatur	4574749147	communty[patch]: Release 0.0.3 (#14673)	9 months ago
Erick Friis	c5250f12c2	cli[patch]: unicode issue (#14672) ... Some operating systems compile template, resulting in unicode decode errors	9 months ago
William FH	75b8891399	Update Vertex AI to include Gemini (#14670) ... h/t to @lkuligin - **Description:** added new models on VertexAI - **Twitter handle:** @lkuligin --------- Co-authored-by: Leonid Kuligin <lkuligin@yandex.ru> Co-authored-by: Harrison Chase <hw.chase.17@gmail.com>	9 months ago
Erick Friis	858f4cbce4	cli[patch]: rc (#14667)	9 months ago
Tomaz Bratanic	ea2616ae23	Fix RRF and lucene escape characters for neo4j vector store (#14646) ... * Remove Lucene special characters (fixes https://github.com/langchain-ai/langchain/issues/14232) * Fixes RRF normalization for hybrid search	9 months ago
Erick Friis	7e6ca3c2b9	cli[patch]: integration template (#14571)	9 months ago
James Braza	b9ef92f2f4	Fixed `DeprecationWarning` for `PromptTemplate.from_file` module-level calls (#14468) ... Resolves https://github.com/langchain-ai/langchain/issues/14467	9 months ago
Chengzu Ou	df95abb7e7	docs: Add Databricks Vector Search example notebook (#14158) ... This PR adds an example notebook for the Databricks Vector Search vector store. It also adds an introduction to the Databricks Vector Search product on the Databricks's provider page. --------- Co-authored-by: Bagatur <baskaryan@gmail.com>	9 months ago
葛尧	e780433f6b	Fix token_usage None issue in ChatOpenAI with local Chatglm2-6B (#14493) ... When using local Chatglm2-6B by changing OPENAI_BASE_URL to localhost, the token_usage in ChatOpenAI becomes None. This leads to an AttributeError when trying to access token_usage.items(). This commit adds a check to ensure token_usage is not None before accessing its items. This change prevents the AttributeError and allows ChatOpenAI to work seamlessly with a local Chatglm2-6B model, aligning with the way it operates with the OpenAI API. <!-- Thank you for contributing to LangChain! Replace this entire comment with: - **Description:** a description of the change, - **Issue:** the issue # it fixes (if applicable), - **Dependencies:** any dependencies required for this change, - **Tag maintainer:** for a quicker response, tag the relevant maintainer (see below), - **Twitter handle:** we announce bigger features on Twitter. If your PR gets announced, and you'd like a mention, we'll gladly shout you out! Please make sure your PR is passing linting and testing before submitting. Run `make format`, `make lint` and `make test` to check this locally. See contribution guidelines for more information on how to write/run tests, lint, etc: https://github.com/langchain-ai/langchain/blob/master/.github/CONTRIBUTING.md If you're adding a new integration, please include: 1. a test for the integration, preferably unit tests that do not rely on network access, 2. an example notebook showing its use. It lives in `docs/extras` directory. If no one reviews your PR within a few days, please @-mention one of @baskaryan, @eyurtsev, @hwchase17. --> Co-authored-by: Harrison Chase <hw.chase.17@gmail.com>	9 months ago
Massimiliano Pronesti	6080c98108	fix(embeddings): huggingface hub embeddings and TEI (#14489) ... **Description:** This PR fixes `HuggingFaceHubEmbeddings` by making the API token optional (as in the client beneath). Most models don't require one. I also updated the notebook for TEI (text-embeddings-inference) accordingly as requested here #14288. In addition, I fixed a mistake in the POST call parameters. **Tag maintainers:** @baskaryan	9 months ago
Thomas B	b4e3e47c92	feat: Yaml output parser (#14496) ... ## Description New YAML output parser as a drop-in replacement for the Pydantic output parser. Yaml is a much more token-efficient format than JSON, proving to be **~35% faster and using the same percentage fewer completion tokens**. ☑️ Formatted ☑️ Linted ☑️ Tested (analogous to the existing`test_pydantic_parser.py`) The YAML parser excels in situations where a list of objects is required, where the root object needs no key: ```python class Products(BaseModel): __root__: list[Product] ``` I ran the prompt `Generate 10 healthy, organic products` 10 times on one chain using the `PydanticOutputParser`, the other one using the`YamlOutputParser` with `Products` (see below) being the targeted model to be created. LLMs used were Fireworks' `lama-v2-34b-code-instruct` and OpenAI `gpt-3.5-turbo`. All runs succeeded without validation errors. ```python class Nutrition(BaseModel): sugar: int = Field(description="Sugar in grams") fat: float = Field(description="% of daily fat intake") class Product(BaseModel): name: str = Field(description="Product name") stats: Nutrition class Products(BaseModel): """A list of products""" products: list[Product] # Used `__root__` for the yaml chain ``` Stats after 10 runs reach were as follows: ### JSON ø time: 7.75s ø tokens: 380.8 ### YAML ø time: 5.12s ø tokens: 242.2 Looking forward to feedback, tips and contributions!	9 months ago
Bob Lin	a019183a01	create mypy cache dir if it doesn't exist (#14579) ... ### Description When running `make lint` multiple times, i can see the error `mkdir: .mypy_cache: File exists`. Use `mkdir -p` to solve this problem. <img width="1512" alt="Screenshot 2023-12-12 at 11 22 01 AM" src="https://github.com/langchain-ai/langchain/assets/10000925/1429383d-3283-4e22-8882-5693bc50b502">	9 months ago
dandanwei	e5bd88383f	fix a bug in RedisNum filter againt value 0 (#14587) ... - **Description:** There is a bug in RedisNum filter that filter towards value 0 will be parsed as "*". This is a fix to it. - **Issue:** NA - **Dependencies:** NA - **Tag maintainer:** NA - **Twitter handle:** NA	9 months ago
Lance Martin	282362382c	Minor update to ensemble retriever to handle a mix of Documents or str (#14552)	9 months ago