Commit Graph

6750 Commits

Author SHA1 Message Date
William FH
8265be4d3e
Unset context to None in var (#30380) 2025-03-19 18:53:17 +00:00
William FH
4130e6476b
Unset context after step (#30378)
While we are already careful to copy before setting the config, if other
objects hold a reference to the config or context, it wouldn't be
cleared.
2025-03-19 11:46:23 -07:00
Vadym Barda
37190881d3
core[patch]: add util for approximate token counting (#30373) 2025-03-19 17:48:38 +00:00
Matthew Farrellee
5f812f5968
langchain-tests: skip instead of passing image message tests (#30375)
**Description:** use skip for image message tests
2025-03-19 15:35:32 +00:00
ccurme
aae8306d6c
groq: release 0.3.0 (#30374) 2025-03-19 15:23:30 +00:00
Ashwin
83cfb9691f
Fix typo: change 'ben' to 'be' in comment (#30358)
**Description:**  
This PR fixes a minor typo in the comments within
`libs/partners/openai/langchain_openai/chat_models/base.py`. The word
"ben" has been corrected to "be" for clarity and professionalism.

**Issue:**  
N/A

**Dependencies:**  
None
2025-03-19 10:35:35 -04:00
Florian Chappaz
07cb41ea9e
community: aligning ChatLiteLLM default parameters with litellm (#30360)
**Description:**
Since `ChatLiteLLM` is forwarding most parameters to
`litellm.completion(...)`, there is no reason to set other default
values than the ones defined by `litellm`.

In the case of parameter 'n', it also provokes an issue when trying to
call a serverless endpoint on Azure, as it is considered an extra
parameter. So we need to keep it optional.

We can debate about backward compatibility of this change: in my
opinion, there should not be big issues since from my experience,
calling `litellm.completion()` without these parameters works fine.

**Issue:** 
- #29679 

**Dependencies:** None
2025-03-19 09:07:28 -04:00
Hodory
57ffacadd0
community: add keep_newlines parameter to process_pages method (#30365)
- **Description:** Adding keep_newlines parameter to process_pages
method with page_ids on Confluence document loader
- **Issue:** N/A (This is an enhancement rather than a bug fix)
- **Dependencies:** N/A
- **Twitter handle:** N/A
2025-03-19 08:57:59 -04:00
William FH
f5a0092551
Rm test for parent_run presence (#30356) 2025-03-18 19:44:19 -07:00
Adam Brenner
f949d9a3d3
docs: Add Dell PowerScale Document Loader (#30209)
# Description
Adds documentation on LangChain website for a Dell specific document
loader for on-prem storage devices. Additional details on what the
document loader is described in the PR as well as on our github repo:
[https://github.com/dell/powerscale-rag-connector](https://github.com/dell/powerscale-rag-connector)

This PR also creates a category on the document loader webpage as no
existing category exists for on-prem. This follows the existing pattern
already established as the website has a category for cloud providers.

# Issue:
New release, no issue.

# Dependencies:

None

# Twitter handle:

DellTech

---------

Signed-off-by: Adam Brenner <adam@aeb.io>
Co-authored-by: Chester Curme <chester.curme@gmail.com>
2025-03-18 22:39:21 -04:00
ccurme
9fb0db6937
community: release 0.3.20 (#30354) 2025-03-18 21:57:12 +00:00
ccurme
168f1dfd93
langchain[patch]: update text-splitters min bound (#30352) 2025-03-18 20:53:43 +00:00
ccurme
f6cf2ce2ad
langchain[patch]: lock with latest text-splitters (#30350) 2025-03-18 19:29:11 +00:00
ccurme
2909b49045
langchain: release 0.3.21 (#30348) 2025-03-18 19:13:20 +00:00
ccurme
958f85d541
text-splitters: release 0.3.7 (#30347) 2025-03-18 19:11:37 +00:00
Lance Martin
46d6bf0330
ollama[minor]: update default method for structured output (#30273)
From function calling to Ollama's [dedicated structured output
feature](https://ollama.com/blog/structured-outputs).

---------

Co-authored-by: Chester Curme <chester.curme@gmail.com>
2025-03-18 12:44:22 -04:00
Marlene
ff8ce60dcc
Core: Adding Azure AI to Supported Chat Models (#30342)
- **Description:** I was testing out `init_chat` and saw that chat
models can now be inferred. Azure OpenAI is currently only supported but
we would like to add support for Azure AI which is a different package.
This PR edits the `base.py` file to add the chat implementation.
- I don't think this adds any additional dependencies 
- Will add a test and lint, but starting an initial draft PR. 

cc @santiagxf

---------

Co-authored-by: Chester Curme <chester.curme@gmail.com>
2025-03-18 11:53:20 -04:00
TheSongg
251551ccf1
doc: Implement langchain-xinference (#30296)
- [ ] **PR title**: Implement langchain-xinference

- [ ] **PR message**: 
Implement a standalone package for Xinference chat models and llm
models.

https://github.com/langchain-ai/langchain/issues/30045#issue-2887214214
2025-03-18 11:50:16 -04:00
wenmeng zhou
5a6e1254a7
support return reasoning content for models like qwq in dashscope (#30317)
Thank you for contributing to LangChain!

- [x] **PR title**: "package: description"
- Where "package" is whichever of langchain, community, core, etc. is
being modified. Use "docs: ..." for purely docs changes, "infra: ..."
for CI changes.
  - Example: "community: add foobar LLM"

here is an example
```python
from langchain_community.chat_models.tongyi import ChatTongyi
from langchain_core.messages import HumanMessage

chatLLM = ChatTongyi(
    model="qwq-32b",   # refer to  https://help.aliyun.com/zh/model-studio/getting-started/models for more models
)
res = chatLLM.stream([HumanMessage(content="how much is 1 plus 1")])
for r in res:
    print(r)
```

```shell
content='' additional_kwargs={'reasoning_content': 'Okay, so the'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' user is asking "'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': 'how much is 1 plus'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' 1." Let me think'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' about this. Hmm'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ', 1 plus'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': " 1... That's a pretty"} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' basic math question. I'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' remember from arithmetic that when'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' you add 1 and'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' 1 together, the'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' result is 2.'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' But wait, maybe'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' I should double-check to be'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' sure. Let me visualize it'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': '. If I have one apple'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' and someone gives me another'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' apple, I have'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' two apples total. Yeah,'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' that makes sense. Or'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' on a number line'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ', starting at 1 and'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' moving 1 step forward lands'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' you at 2'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': '. \n\nIs there any'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' context where 1 +'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' 1 might not equal'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' 2? Like in different'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' number bases? Let'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': "'s see. In base"} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' 10, which'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' is standard,'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' 1+1 is'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' 2. But if'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' we were in binary'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' (base 2'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': '), 1 +'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' 1 would be 1'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': '0. But the question'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': " doesn't specify a base,"} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' so I think the'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' default is base 10'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': '. \n\nAlternatively, could'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' this be a trick'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' question? Maybe they'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': "'re referring to something else"} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ', like in Boolean'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' algebra where 1 +'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' 1 might still'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' be 1 in'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' some contexts? Wait'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ', no, in Boolean'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' addition, 1'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' + 1 is typically'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': " 1 because it's logical"} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' OR. But the'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' question just says "1'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' plus 1," which is'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' more arithmetic than Boolean.'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' \n\nOr maybe in some other'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' mathematical structure like modular arithmetic?'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' For example, modulo'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' 2,'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' 1 + 1 is'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' 0. But again'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ', unless specified, it'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': "'s probably standard addition"} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': '. \n\nThe user might be'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' testing if I know basic'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' math, or maybe'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': " they're a student just"} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' starting out. Either way,'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' the straightforward answer is'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' 2. I should also'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': " consider if there's any cultural"} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' references or jokes where'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' 1 + 1 equals'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' something else, but I can'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': "'t think of any common"} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' ones. \n\nAlternatively'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ', in some contexts like'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' in chemistry,'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' 1 + 1 could refer'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' to mixing solutions, but that'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': "'s not standard. The question"} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' is pretty simple,'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' so I think the answer'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' is 2. To'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' be thorough, maybe mention'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' that in standard arithmetic it'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': "'s 2, but if"} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': " there's a different"} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' context, the answer'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' might vary. But since'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' no context is given'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ', 2 is the safest'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ' answer.'} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='The result' additional_kwargs={'reasoning_content': ''} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content=' of 1 plus' additional_kwargs={'reasoning_content': ''} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content=' 1 is **2**.' additional_kwargs={'reasoning_content': ''} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content=' \n\nIn standard arithmetic (base' additional_kwargs={'reasoning_content': ''} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content=' 10), adding' additional_kwargs={'reasoning_content': ''} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content=' 1 and 1 together' additional_kwargs={'reasoning_content': ''} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content=' yields 2. This is' additional_kwargs={'reasoning_content': ''} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content=' a fundamental mathematical principle. If' additional_kwargs={'reasoning_content': ''} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content=' the question involves a different context' additional_kwargs={'reasoning_content': ''} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content=' (e.g., binary' additional_kwargs={'reasoning_content': ''} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content=', modular arithmetic, or a' additional_kwargs={'reasoning_content': ''} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content=' metaphorical meaning), it' additional_kwargs={'reasoning_content': ''} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content=' would need clarification,' additional_kwargs={'reasoning_content': ''} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content=' but under typical circumstances, the' additional_kwargs={'reasoning_content': ''} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content=' answer is **2**.' additional_kwargs={'reasoning_content': ''} response_metadata={} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'
content='' additional_kwargs={'reasoning_content': ''} response_metadata={'finish_reason': 'stop', 'request_id': '4738c641-6bd8-9efc-a4fe-d929d4e62bef', 'token_usage': {'input_tokens': 16, 'output_tokens': 560, 'total_tokens': 576}} id='run-bd026918-16e5-429f-aa75-3ff7701e9f8d'

```

Co-authored-by: ccurme <chester.curme@gmail.com>
2025-03-18 11:43:10 -04:00
ccurme
b91daf06eb
groq[minor]: remove default model (#30341)
The default model for `ChatGroq`, `"mixtral-8x7b-32768"`, is being
retired on March 20, 2025. Here we remove the default, such that model
names must be explicitly specified (being explicit is a good practice
here, and avoids the need for breaking changes down the line). This
change will be released in a minor version bump to 0.3.

This follows https://github.com/langchain-ai/langchain/pull/30161
(released in version 0.2.5), where we began generating warnings to this
effect.

![Screenshot 2025-03-18 at 10 33
27 AM](https://github.com/user-attachments/assets/f1e4b302-c62a-43b0-aa86-eaf9271e86cb)
2025-03-18 10:50:34 -04:00
amuwall
f6a17fbc56
community: fix import exception too constrictive (#30218)
Fix this issue #30097
2025-03-17 22:09:02 -04:00
qonnop
036f00dc92
community: support in-memory data (Blob.from_data) in all audio parsers (#30262)
OpenAIWhisperParser, OpenAIWhisperParserLocal, YandexSTTParser do not
handle in-memory audio data (loaded via Blob.from_data) correctly. They
require Blob.path to be set and AudioSegment is always read from the
file system. In-memory data is handled correctly only for
FasterWhisperParser so far. I changed OpenAIWhisperParser,
OpenAIWhisperParserLocal, YandexSTTParser accordingly to match
FasterWhisperParser.
Thanks for reviewing the PR!

Co-authored-by: qonnop <qonnop@users.noreply.github.com>
2025-03-17 19:52:33 -04:00
Matthew Farrellee
1985aaf095
langchain-tests: allow subclasses to add addition, non-standard tests (#30204)
**description:** the ChatModel[Integration]Tests classes are powerful
and helpful, this change allows sub-classes to add additional tests.

for instance,

```
class TestChatMyServiceIntegration(ChatModelIntegrationTests):
    ...
    def test_myservice(self, model: BaseChatModel) -> None:
        ...
```

---------

Co-authored-by: ccurme <chester.curme@gmail.com>
2025-03-17 23:37:16 +00:00
Ben
789db7398b
text-splitters: Add JSFrameworkTextSplitter for Handling JavaScript Framework Code (#28972)
## Description
This pull request introduces a new text splitter,
`JSFrameworkTextSplitter`, to the Langchain library. The
`JSFrameworkTextSplitter` extends the `RecursiveCharacterTextSplitter`
to handle JavaScript framework code effectively, including React (JSX),
Vue, and Svelte. It identifies and utilizes framework-specific component
tags and syntax elements as splitting points, alongside standard
JavaScript syntax. This ensures that code is divided at natural
boundaries, enhancing the parsing and processing of JavaScript and
framework-specific code.

### Key Features
- Supports React (JSX), Vue, and Svelte frameworks.
- Identifies and uses framework-specific tags and syntax elements as
natural splitting points.
- Extends the existing `RecursiveCharacterTextSplitter` for seamless
integration.

## Issue
No specific issue addressed.

## Dependencies
No additional dependencies required.

---------

Co-authored-by: ccurme <chester.curme@gmail.com>
2025-03-17 23:32:33 +00:00
ccurme
5684653775
openai[patch]: release 0.3.9 (#30325) 2025-03-17 16:08:41 +00:00
ccurme
eb9b992aa6
openai[patch]: support additional Responses API features (#30322)
- Include response headers
- Max tokens
- Reasoning effort
- Fix bug with structured output / strict
- Fix bug with simultaneous tool calling + structured output
2025-03-17 12:02:21 -04:00
Bae-ChangHyun
d8510270ee
community: add 'extract' mode to FireCrawlLoader for structured data extraction (#30242)
**Description:** 
Added an 'extract' mode to FireCrawlLoader that enables structured data
extraction from web pages. This feature allows users to Extract
structured data from a single URLs, or entire websites using Large
Language Models (LLMs).
You can show more params and usage on [firecrawl
docs](https://docs.firecrawl.dev/features/extract-beta).
You can extract from only one url now.(it depends on firecrawl's extract
method)

**Dependencies:** 
No new dependencies required. Uses existing FireCrawl API capabilities.

---------

Co-authored-by: chbae <chbae@gcsc.co.kr>
Co-authored-by: ccurme <chester.curme@gmail.com>
2025-03-17 15:15:57 +00:00
qonnop
747efa16ec
community: fix CPU support for FasterWhisperParser (implicit compute type for WhisperModel) (#30263)
FasterWhisperParser fails on a machine without an NVIDIA GPU: "Requested
float16 compute type, but the target device or backend do not support
efficient float16 computation." This problem arises because the
WhisperModel is called with compute_type="float16", which works only for
NVIDIA GPU.

According to the [CTranslate2
docs](https://opennmt.net/CTranslate2/quantization.html#bit-floating-points-float16)
float16 is supported only on NVIDIA GPUs. Removing the compute_type
parameter solves the problem for CPUs. According to the [CTranslate2
docs](https://opennmt.net/CTranslate2/quantization.html#quantize-on-model-loading)
setting compute_type to "default" (standard when omitting the parameter)
uses the original compute type of the model or performs implicit
conversion for the specific computation device (GPU or CPU). I suggest
to remove compute_type="float16".

@hulitaitai you are the original author of the FasterWhisperParser - is
there a reason for setting the parameter to float16?

Thanks for reviewing the PR!

Co-authored-by: qonnop <qonnop@users.noreply.github.com>
2025-03-14 22:22:29 -04:00
ccurme
c74e7b997d
openai[patch]: support structured output via Responses API (#30265)
Also runs all standard tests using Responses API.
2025-03-14 15:14:23 -04:00
Priyansh Agrawal
f54f14b747
community: cube document loader - do not load non-public dimensions and measures (#30286)
Thank you for contributing to LangChain!

- [x] **PR title**: "package: description"
- Where "package" is whichever of langchain, community, core, etc. is
being modified. Use "docs: ..." for purely docs changes, "infra: ..."
for CI changes.
  - Example: "community: add foobar LLM"

- **Description:** Do not load non-public dimensions and measures
(public: false) with Cube semantic loader

- **Issue:** Currently, non-public dimensions and measures are loaded by
the Cube document loader which leads to downstream applications using
these which is not allowed by Cube.


- [x] **Add tests and docs**: 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/docs/integrations` directory.


- [x] **Lint and test**: Run `make format`, `make lint` and `make test`
from the root of the package(s) you've modified. See contribution
guidelines for more: https://python.langchain.com/docs/contributing/

Additional guidelines:
- Make sure optional dependencies are imported within a function.
- Please do not add dependencies to pyproject.toml files (even optional
ones) unless they are required for unit tests.
- Most PRs should not touch more than one package.
- Changes should be backwards compatible.
- If you are adding something to community, do not re-import it in
langchain.

If no one reviews your PR within a few days, please @-mention one of
baskaryan, eyurtsev, ccurme, vbarda, hwchase17.
2025-03-14 15:07:56 -04:00
Stavros Kontopoulos
ac22cde130
langchain_ollama: Support keep_alive in embeddings (#30251)
- Description: Adds support for keep_alive in Ollama Embeddings see
https://github.com/ollama/ollama/issues/6401.
Builds on top of of
https://github.com/langchain-ai/langchain/pull/29296. I have this use
case where I want to keep the embeddings model in cpu forever.
- Dependencies: no deps are being introduced.
- Issue: haven't created an issue yet.
2025-03-14 14:56:50 -04:00
homeffjy
2c99f12062
community[patch]: fix bilibili loader handling of multi-page content (#30283)
Previously the loader would only extract subtitles from the first page
of multi-page videos.
2025-03-14 14:53:03 -04:00
ccurme
d5d0134e7b
anthropic: release 0.3.10 (#30287) 2025-03-14 16:23:21 +00:00
ccurme
226f29bc96
anthropic: support built-in tools, improve docs (#30274)
- Support features from recent update:
https://www.anthropic.com/news/token-saving-updates (mostly adding
support for built-in tools in `bind_tools`
- Add documentation around prompt caching, token-efficient tool use, and
built-in tools.
2025-03-14 16:18:50 +00:00
Priyansh Agrawal
f27e2d7ce7
community: cube document loader - fix logging (#30285)
Thank you for contributing to LangChain!

- [x] **PR title**: "package: description"
- Where "package" is whichever of langchain, community, core, etc. is
being modified. Use "docs: ..." for purely docs changes, "infra: ..."
for CI changes.
  - Example: "community: add foobar LLM"

- **Description:** Fix bad log message on line#56 and replace f-string
logs with format specifiers

- **Issue:** Log messages such as this one
`INFO:langchain_community.document_loaders.cube_semantic:Loading
dimension values for: {dimension_name}...`

- [ ] **Add tests and docs**: 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/docs/integrations` directory.


- [x] **Lint and test**: Run `make format`, `make lint` and `make test`
from the root of the package(s) you've modified. See contribution
guidelines for more: https://python.langchain.com/docs/contributing/

Additional guidelines:
- Make sure optional dependencies are imported within a function.
- Please do not add dependencies to pyproject.toml files (even optional
ones) unless they are required for unit tests.
- Most PRs should not touch more than one package.
- Changes should be backwards compatible.
- If you are adding something to community, do not re-import it in
langchain.

If no one reviews your PR within a few days, please @-mention one of
baskaryan, eyurtsev, ccurme, vbarda, hwchase17.
2025-03-14 11:36:18 -04:00
ccurme
bbd4b36d76
mistralai[patch]: bump core (#30278) 2025-03-13 23:04:36 +00:00
ccurme
315bb17ef5
core: release 0.3.45 (#30277) 2025-03-13 22:44:23 +00:00
pulvedu
d0bfc7f820
community[fix] : Pass API_KEY as argument (#30272)
PR Title:
community: Fix Pass API_KEY as argument

PR Message:
Description:
This PR fixes validation error "Value error, Did not find
tavily_api_key, please add an environment variable `TAVILY_API_KEY`
which contains it, or pass `tavily_api_key` as a named parameter."

Dependencies:
No new dependencies introduced.

---------

Co-authored-by: pulvedu <dustin@tavily.com>
Co-authored-by: Chester Curme <chester.curme@gmail.com>
2025-03-13 22:19:31 +00:00
ccurme
733abcc884
mistral: release 0.2.8 (#30275) 2025-03-13 21:54:34 +00:00
Jacob Lee
e9c1765967
fix(core): Ignore missing secrets on deserialization (#30252) 2025-03-13 12:27:03 -07:00
ccurme
ebea5e014d
standard tests: test simple agent loop (#30268) 2025-03-13 16:34:12 +00:00
ccurme
cd1ea8e94d
openai[patch]: support Responses API (#30231)
Co-authored-by: Bagatur <baskaryan@gmail.com>
2025-03-12 12:25:46 -04:00
Jason Zhang
49bdd3b6fe
docs: Add AgentQL provider doc, tool/toolkit doc and documentloader doc (#30144)
- **Description:** Added AgentQL docs for the provider page, tools page
and documentloader page
- **Twitter handle:** @AgentQL

Repo:
https://github.com/tinyfish-io/agentql-integrations/tree/main/langchain
PyPI: https://pypi.org/project/langchain-agentql/

If no one reviews your PR within a few days, please @-mention one of
baskaryan, eyurtsev, ccurme, vbarda, hwchase17.

---------

Co-authored-by: Chester Curme <chester.curme@gmail.com>
2025-03-11 21:57:40 -04:00
Vadym Barda
23fa70f328
core[patch]: release 0.3.44 (#30236) 2025-03-11 18:59:02 -04:00
Vadym Barda
c7842730ef
core[patch]: support single-node subgraphs and put subgraph nodes under the respective subgraphs (#30234) 2025-03-11 18:55:45 -04:00
ccurme
62c570dd77
standard-tests, openai: bump core (#30202) 2025-03-10 19:22:24 +00:00
ccurme
f896e701eb
deepseek: install local langchain-tests in test deps (#30198) 2025-03-10 16:58:17 +00:00
Hugh Gao
aa6dae4a5b
community: Remove the system message count limit for ChatTongyi. (#30192)
## Description
The models in DashScope support multiple SystemMessage. Here is the
[Doc](https://bailian.console.aliyun.com/model_experience_center/text#/model-market/detail/qwen-long?tabKey=sdk),
and the example code on the document page:
```python
import os
from openai import OpenAI

client = OpenAI(
    api_key=os.getenv("DASHSCOPE_API_KEY"),  # 如果您没有配置环境变量,请在此处替换您的API-KEY
    base_url="https://dashscope.aliyuncs.com/compatible-mode/v1",  # 填写DashScope服务base_url
)
# 初始化messages列表
completion = client.chat.completions.create(
    model="qwen-long",
    messages=[
        {'role': 'system', 'content': 'You are a helpful assistant.'},
        # 请将 'file-fe-xxx'替换为您实际对话场景所使用的 file-id。
        {'role': 'system', 'content': 'fileid://file-fe-xxx'},
        {'role': 'user', 'content': '这篇文章讲了什么?'}
    ],
    stream=True,
    stream_options={"include_usage": True}
)

full_content = ""
for chunk in completion:
    if chunk.choices and chunk.choices[0].delta.content:
        # 拼接输出内容
        full_content += chunk.choices[0].delta.content
        print(chunk.model_dump())

print({full_content})
```
Tip: The example code is for OpenAI, but the document said that it also
supports the DataScope API, and I tested it, and it works.
```
Is the Dashscope SDK invocation method compatible?

Yes, the Dashscope SDK remains compatible for model invocation. However, file uploads and file-ID retrieval are currently only supported via the OpenAI SDK. The file-ID obtained through this method is also compatible with Dashscope for model invocation.
```
2025-03-10 08:58:40 -04:00
ccurme
67aff1648b
community: Add OpenGradient integration (Toolkit) (#30190)
Commandeering https://github.com/langchain-ai/langchain/pull/30135

---------

Co-authored-by: kylexqian <kylexqian@gmail.com>
2025-03-09 18:08:07 -04:00
ccurme
b209d46eb3
mistral[patch]: set global ssl context (#30189) 2025-03-09 21:27:41 +00:00
Vijay Selvaraj
df459d0d5e
community: add Valthera integration (#30105)
```markdown
**Description:**  
This PR integrates Valthera into LangChain, introducing an framework designed to send highly personalized nudges by an LLM agent. This is modeled after Dr. BJ Fogg's Behavior Model. This integration includes:

- Custom data connectors for HubSpot, PostHog, and Snowflake.
- A unified data aggregator that consolidates user data.
- Scoring configurations to compute motivation and ability scores.
- A reasoning engine that determines the appropriate user action.
- A trigger generator to create personalized messages for user engagement.

**Issue:**  
N/A

**Dependencies:**  
N/A

**Twitter handle:**  
- `@vselvarajijay`

**Tests and Docs:**  
- `docs/docs/integrations/tools/valthera` 
- `https://github.com/valthera/langchain-valthera/tree/main/tests`

```

---------

Co-authored-by: Chester Curme <chester.curme@gmail.com>
2025-03-09 21:19:08 +00:00
ccurme
3823daa0b9
cli: update integration doc template for tools (#30188)
Chain example -> langgraph agent
2025-03-09 21:14:43 +00:00
Jonathan Feng
911accf733
docs: add contextualai documentation (#30050)
Thank you for contributing to LangChain!
 
**Description:** adds ContextualAI's `langchain-contextual` package's
documentation

If no one reviews your PR within a few days, please @-mention one of
baskaryan, efriis, eyurtsev, ccurme, vbarda, hwchase17.

---------

Co-authored-by: Chester Curme <chester.curme@gmail.com>
2025-03-09 02:43:13 +00:00
Bharat
b9746a6910
fixes#30182: update tool names to match OpenAI function name pattern (#30183)
The OpenAI API requires function names to match the pattern
'^[a-zA-Z0-9_-]+$'. This updates the JIRA toolkit's tool names to use
underscores instead of spaces to comply with this requirement and
prevent BadRequestError when using the tools with OpenAI functions.

Error fixed:
```
File "langgraph-bug-fix/.venv/lib/python3.13/site-packages/openai/_base_client.py", line 1023, in _request
    raise self._make_status_error_from_response(err.response) from None
openai.BadRequestError: Error code: 400 - {'error': {'message': "Invalid 'tools[0].function.name': string does not match pattern. Expected a string that matches the pattern '^[a-zA-Z0-9_-]+$'.", 'type': 'invalid_request_error', 'param': 'tools[0].function.name', 'code': 'invalid_value'}}
During task with name 'agent' and id 'aedd7537-e8d5-6678-d0c5-98129586d3ac'
```

Issue:#30182
2025-03-08 20:48:25 -05:00
ccurme
cee0fecb08
docs: update package registry counts (#30181) 2025-03-08 20:37:59 -05:00
William FH
bac3a28e70
Flush (#30157) 2025-03-07 16:32:15 -08:00
ccurme
a7ab5e8372
community[patch]: ChatPerplexity: track usage metadata (#30175) 2025-03-07 23:25:05 +00:00
ccurme
1c993b921c
core[patch]: release 0.3.43 (#30173) 2025-03-07 21:56:00 +00:00
ccurme
9893e5cb80
core[patch]: catch structured_output_format (#30172)
Change to `ls_structured_output_format` was not backward-compatible with
older versions of integration packages.
2025-03-07 16:50:06 -05:00
ccurme
33a3510243
core[patch]: export ArgsSchema (#30169)
This is needed for type hints

see: https://github.com/langchain-ai/langchain/pull/30167
2025-03-07 20:43:05 +00:00
ccurme
17507c9ba6
groq[patch]: release 0.2.5 (#30168) 2025-03-07 20:25:51 +00:00
andyzhou1982
9e863c89d2
add JiebaLinkExtractor for chinese doc extracting (#30150)
Thank you for contributing to LangChain!

- [ ] **PR title**: "community: chinese doc extracting"


- [ ] **PR message**: 
- **Description:** add jieba_link_extractor.py for chinese doc
extracting
    - **Dependencies:** jieba


- [ ] **Add tests and docs**: If you're adding a new integration, please
include
  /doc/doc/integrations/providers/jieba.md
  /doc/doc/integrations/vectorstores/jieba_link_extractor.ipynb
  /libs/packages.yml

---------

Co-authored-by: Chester Curme <chester.curme@gmail.com>
2025-03-07 20:21:46 +00:00
ccurme
74e7772a5f
groq[patch]: warn if model is not specified (#30161)
Groq is retiring `mixtral-8x7b-32768`, which is currently the default
model for ChatGroq, on March 20. Here we emit a warning if the model is
not specified explicitly.

A version 0.3.0 will be released ahead of March 20 that removes the
default altogether.
2025-03-07 15:21:13 -05:00
Ioannis Bakagiannis
3444e587ee
docs: Integration Update - ADS4GPTs (#30153)
docs: New integration for LangChain - ads4gpts-langchain

Description: Tools and Toolkit for Agentic integration natively within
LangChain with ADS4GPTs, in order to help applications monetize with
advertising.

Twitter handle: @ads4gpts

Co-authored-by: knitlydevaccount <loom+github@knitly.app>
2025-03-07 14:35:44 -05:00
ccurme
3c258194ae
tests[patch]: release 0.3.14 (#30165) 2025-03-07 18:34:05 +00:00
ccurme
34638ccfae
openai[patch]: release 0.3.8 (#30164) 2025-03-07 18:26:40 +00:00
ccurme
4e5058f29c
core[patch]: release 0.3.42 (#30163) 2025-03-07 18:14:45 +00:00
Eugene Yurtsev
894fd63a61
cli: release 0.0.36 (#30159)
Bump for 0.0.36
2025-03-07 13:05:40 -05:00
ccurme
806211475a
core[patch]: update structured output tracing (#30123)
- Trace JSON schema in `options`
- Rename to `ls_structured_output_format`
2025-03-07 13:05:25 -05:00
ccurme
230876a7c5
anthropic[patch]: add PDF input example to API reference (#30156) 2025-03-07 14:19:08 +00:00
joeconstantino
022ff9eead
Tableau docs for new datasource qa tool (#30125)
- **Description: a notebook showing langchain and langraph agents using
the new langchain_tableau tool
- **Twitter handle: @joe_constantin0

---------

Co-authored-by: Joe Constantino <joe@constantino.com>
Co-authored-by: Chester Curme <chester.curme@gmail.com>
2025-03-06 14:58:56 +00:00
ccurme
52b0570bec
core, openai, standard-tests: improve OpenAI compatibility with Anthropic content blocks (#30128)
- Support thinking blocks in core's `convert_to_openai_messages` (pass
through instead of error)
- Ignore thinking blocks in ChatOpenAI (instead of error)
- Support Anthropic-style image blocks in ChatOpenAI

---

Standard integration tests include a `supports_anthropic_inputs`
property which is currently enabled only for tests on `ChatAnthropic`.
This test enforces compatibility with message histories of the form:
```
- system message
- human message
- AI message with tool calls specified only through `tool_use` content blocks
- human message containing `tool_result` and an additional `text` block
```
It additionally checks support for Anthropic-style image inputs if
`supports_image_inputs` is enabled.

Here we change this test, such that if you enable
`supports_anthropic_inputs`:
- You support AI messages with text and `tool_use` content blocks
- You support Anthropic-style image inputs (if `supports_image_inputs`
is enabled)
- You support thinking content blocks.

That is, we add a test case for thinking content blocks, but we also
remove the requirement of handling tool results within HumanMessages
(motivated by existing agent abstractions, which should all return
ToolMessage). We move that requirement to a ChatAnthropic-specific test.
2025-03-06 09:53:14 -05:00
Pat Patterson
b3dc66f7a3
community: fix AttributeError when creating LanceDB vectorstore (#30127)
**Description:**

This PR adds a call to `guard_import()` to fix an AttributeError raised
when creating LanceDB vectorstore instance with an existing LanceDB
table.

**Issue:**

This PR fixes issue #30124.

**Dependencies:**

No additional dependencies.

**Twitter handle:**

[@metadaddy](https://x.com/metadaddy), but I spend more time at
[@metadaddy.net](https://bsky.app/profile/metadaddy.net) these days.

---------

Co-authored-by: Chester Curme <chester.curme@gmail.com>
2025-03-05 23:04:38 +00:00
Hugh Gao
9b7b8e4a1a
community: make DashScope models support Partial Mode for text continuation. (#30108)
## Description
make DashScope models support Partial Mode for text continuation.

For text continuation in ChatTongYi, it supports text continuation with
a prefix by adding a "partial" argument in AIMessage. The document is
[Partial Mode
](https://help.aliyun.com/zh/model-studio/user-guide/partial-mode?spm=a2c4g.11186623.help-menu-2400256.d_1_0_0_8.211e5b77KMH5Pn&scm=20140722.H_2862210._.OR_help-T_cn~zh-V_1).
The API example is:
```py
import os
import dashscope

messages = [{
    "role": "user",
    "content": "请对“春天来了,大地”这句话进行续写,来表达春天的美好和作者的喜悦之情"
},
{
    "role": "assistant",
    "content": "春天来了,大地",
    "partial": True
}]
response = dashscope.Generation.call(
    api_key=os.getenv("DASHSCOPE_API_KEY"),
    model='qwen-plus',
    messages=messages,
    result_format='message',  
)

print(response.output.choices[0].message.content)
```

---------

Co-authored-by: Chester Curme <chester.curme@gmail.com>
2025-03-05 16:22:14 +00:00
黑牛
f0153414d5
Add request_id field to improve request tracking and debugging (for Tongyi model) (#30110)
- **Description**: Added the request_id field to the check_response
function to improve request tracking and debugging, applicable for the
Tongyi model.
- **Issue**: None
- **Dependencies**: None
- **Twitter handle**: None

- **Add tests and docs**: None

- **Lint and test**: Ran `make format`, `make lint`, and `make test` to
ensure the code meets formatting and testing requirements.
2025-03-05 11:03:47 -05:00
Manthan Surkar
1ee8aceaee
community: fix Jira API wrapper failing initialization with cloud param (#30117)
### **Description**  
Converts the boolean `jira_cloud` parameter in the Jira API Wrapper to a
string before initializing the Jira Client. Also adds tests for the
same.

### **Issue**  
[Jira API Wrapper
Bug](8abb65e138/libs/community/langchain_community/utilities/jira.py (L47))

```python
jira_cloud_str = get_from_dict_or_env(values, "jira_cloud", "JIRA_CLOUD")
jira_cloud = jira_cloud_str.lower() == "true"
```

The above code has a bug where the value of `"jira_cloud"` is a boolean.
If it is passed, calling `.lower()` on a boolean raises an error.
Additionally, `False` cannot be passed explicitly since
`get_from_dict_or_env` falls back to environment variables.

Relevant code in `langchain_core`:  

[Source](https://github.com/thesmallstar/langchain/blob/master/.venv/lib/python3.13/site-packages/langchain_core/utils/env.py#L46)

```python
if isinstance(key, str) and key in data and data[key]:  # Here, data[key] is False
```

This PR fixes both issues.

### **Twitter Handle**  
[Manthan Surkar](https://x.com/manthan_surkar)
2025-03-05 10:49:25 -05:00
Adrián Panella
c599ba47d5
core(mermaid): fix error when 3+ subgraph levels (#29970) 2025-03-04 13:27:49 -05:00
Alexander Henlein
417efa30a6
docs: add Taiga Tool integration docs (#30042)
This PR adds documentation for the langchain-taiga Tool integration,
including an example notebook at
'docs/docs/integrations/tools/taiga.ipynb' and updates to
'libs/packages.yml' to track the new package.

Issue:
N/A

Dependencies:
None

Twitter handle:
N/A

---------

Co-authored-by: ccurme <chester.curme@gmail.com>
2025-03-04 17:51:20 +00:00
Mathias Marciano
5f0102242a
Fixed an issue with the OpenAI Assistant's 'retrieval' tool and adding support for the 'attachments' parameter (#30006)
PR Title:
langchain: add attachments support in OpenAIAssistantRunnable

PR Description:
This PR fixes an issue with the "retrieval" tool (internally named
"file_search") in the OpenAI Assistant by adding support for the
"attachments" parameter in the invoke method. This change allows files
to be linked to messages when they are inserted into threads, which is
essential for utilizing OpenAI's Retrieval Augmented Generation (RAG)
feature.

Issue:
N/A

Dependencies:
None

Twitter handle:
N/A

---------

Co-authored-by: Chester Curme <chester.curme@gmail.com>
2025-03-04 17:34:11 +00:00
Philippe PRADOS
4710c1fa8c
community[minor]: Fix regular expression in visualize and outlines modules. (#30002)
Fix invalid escape characteres
2025-03-04 12:23:48 -05:00
ccurme
577c0d0715
community[patch]: release 0.3.19 (#30104) 2025-03-04 16:12:03 +00:00
ccurme
ba5ddb218f
anthropic[patch]: release 0.3.9 (#30103) 2025-03-04 10:53:55 -05:00
ccurme
9383a0536a
tests[patch]: release 0.3.13 (#30102) 2025-03-04 10:53:43 -05:00
ccurme
fb16c25920
langchain[patch]: release 0.3.20 (#30101) 2025-03-04 15:47:27 +00:00
ccurme
692a68bf1c
core[patch]: release 0.3.41 (#30100) 2025-03-04 15:08:57 +00:00
ccurme
484d945500
community[patch]: remove numpy cap for python < 3.12 (#30084) 2025-03-04 09:46:41 -05:00
ZhangShenao
8575d7491f
[Doc] Improve api doc (#30073)
- Update api_doc for `BaseMessage`
- add static method decorator for `retry_runnable`
2025-03-04 09:39:07 -05:00
Samuel Dion-Girardeau
ccb64e9f4f
docs: Fix typo in code samples for max_tokens_for_prompt (#30088)
- **Description:** Fix typo in code samples for max_tokens_for_prompt.
Code blocks had singular "token" but the method has plural "tokens".
- **Issue:** N/A
- **Dependencies:** N/A
- **Twitter handle:** N/A
2025-03-04 09:11:21 -05:00
ArrayPD
c671d54c6f
core: make with_alisteners() example workable. (#30059)
**Description:**

5 fix of example from function with_alisteners() in
libs/core/langchain_core/runnables/base.py
Replace incoherent example output with workable example's output.

1. SyntaxError: unterminated string literal
    print(f"on start callback starts at {format_t(time.time())}
    correct as
    print(f"on start callback starts at {format_t(time.time())}")

2. SyntaxError: unterminated string literal
    print(f"on end callback starts at {format_t(time.time())}
    correct as
    print(f"on end callback starts at {format_t(time.time())}")

3. NameError: name 'Runnable' is not defined
    Fix as
    from langchain_core.runnables import Runnable

4. NameError: name 'asyncio' is not defined
    Fix as
    import asyncio

5. NameError: name 'format_t' is not defined.
    Implement format_t() as
    from datetime import datetime, timezone

    def format_t(timestamp: float) -> str:
return datetime.fromtimestamp(timestamp, tz=timezone.utc).isoformat()
2025-03-01 15:39:02 -05:00
cold-eye
7c175e3fda
Update ascend.py (#30060)
add batch_size to fix oom when embed large amount texts

Thank you for contributing to LangChain!

- [ ] **PR title**: "package: description"
- Where "package" is whichever of langchain, community, core, etc. is
being modified. Use "docs: ..." for purely docs changes, "infra: ..."
for CI changes.
  - Example: "community: add foobar LLM"


- [ ] **PR message**: ***Delete this entire checklist*** and replace
with
    - **Description:** a description of the change
    - **Issue:** the issue # it fixes, if applicable
    - **Dependencies:** any dependencies required for this change
- **Twitter handle:** if your PR gets announced, and you'd like a
mention, we'll gladly shout you out!


- [ ] **Add tests and docs**: 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/docs/integrations` directory.


- [ ] **Lint and test**: Run `make format`, `make lint` and `make test`
from the root of the package(s) you've modified. See contribution
guidelines for more: https://python.langchain.com/docs/contributing/

Additional guidelines:
- Make sure optional dependencies are imported within a function.
- Please do not add dependencies to pyproject.toml files (even optional
ones) unless they are required for unit tests.
- Most PRs should not touch more than one package.
- Changes should be backwards compatible.
- If you are adding something to community, do not re-import it in
langchain.

If no one reviews your PR within a few days, please @-mention one of
baskaryan, efriis, eyurtsev, ccurme, vbarda, hwchase17.
2025-03-01 14:10:41 -05:00
ccurme
3b066dc005
anthropic[patch]: allow structured output when thinking is enabled (#30047)
Structured output will currently always raise a BadRequestError when
Claude 3.7 Sonnet's `thinking` is enabled, because we rely on forced
tool use for structured output and this feature is not supported when
`thinking` is enabled.

Here we:
- Emit a warning if `with_structured_output` is called when `thinking`
is enabled.
- Raise `OutputParserException` if no tool calls are generated.

This is arguably preferable to raising an error in all cases.

```python
from langchain_anthropic import ChatAnthropic
from pydantic import BaseModel


class Person(BaseModel):
    name: str
    age: int


llm = ChatAnthropic(
    model="claude-3-7-sonnet-latest",
    max_tokens=5_000,
    thinking={"type": "enabled", "budget_tokens": 2_000},
)
structured_llm = llm.with_structured_output(Person)  # <-- this generates a warning
```

```python
structured_llm.invoke("Alice is 30.")  # <-- works
```

```python
structured_llm.invoke("Hello!")  # <-- raises OutputParserException
```
2025-02-28 14:44:11 -05:00
ccurme
f8ed5007ea
anthropic, mistral: return model_name in response metadata (#30048)
Took a "census" of models supported by init_chat_model-- of those that
return model names in response metadata, these were the only two that
had it keyed under `"model"` instead of `"model_name"`.
2025-02-28 18:56:05 +00:00
Christophe Bornet
9e6ffd1264
core: Add ruff rules PTH (pathlib) (#29338)
See https://docs.astral.sh/ruff/rules/#flake8-use-pathlib-pth

Co-authored-by: ccurme <chester.curme@gmail.com>
2025-02-28 13:22:20 -05:00
TheSongg
86b364de3b
Add asynchronous generate interface (#30001)
- [ ] **PR title**: [langchain_community.llms.xinference]: Add
asynchronous generate interface

- [ ] **PR message**: The asynchronous generate interface support stream
data and non-stream data.
          
        chain = prompt | llm
        async for chunk in chain.astream(input=user_input):
            yield chunk


- [ ] **Add tests and docs**:

       from langchain_community.llms import Xinference
       from langchain.prompts import PromptTemplate

       llm = Xinference(
server_url="http://0.0.0.0:9997", # replace your xinference server url
model_uid={model_uid} # replace model_uid with the model UID return from
launching the model
           stream = True
            )
prompt = PromptTemplate(input=['country'], template="Q: where can we
visit in the capital of {country}? A:")
       chain = prompt | llm
       async for chunk in chain.astream(input=user_input):
           yield chunk
2025-02-28 12:32:44 -05:00
Fakai Zhao
f07338d2bf
Implementing the MMR algorithm for OLAP vector storage (#30033)
Thank you for contributing to LangChain!

-  **Implementing the MMR algorithm for OLAP vector storage**: 
  - Support Apache Doris and StarRocks OLAP database.
- Example: "vectorstore.as_retriever(search_type="mmr",
search_kwargs={"k": 10})"


- **Implementing the MMR algorithm for OLAP vector storage**: 
    - **Apache Doris
    - **StarRocks
    - **Dependencies:** any dependencies required for this change
- **Twitter handle:** if your PR gets announced, and you'd like a
mention, we'll gladly shout you out!


- **Add tests and docs**: 
- Example: "vectorstore.as_retriever(search_type="mmr",
search_kwargs={"k": 10})"


- [ ] **Lint and test**: Run `make format`, `make lint` and `make test`
from the root of the package(s) you've modified. See contribution
guidelines for more: https://python.langchain.com/docs/contributing/

Additional guidelines:
- Make sure optional dependencies are imported within a function.
- Please do not add dependencies to pyproject.toml files (even optional
ones) unless they are required for unit tests.
- Most PRs should not touch more than one package.
- Changes should be backwards compatible.
- If you are adding something to community, do not re-import it in
langchain.

If no one reviews your PR within a few days, please @-mention one of
baskaryan, efriis, eyurtsev, ccurme, vbarda, hwchase17.

---------

Co-authored-by: fakzhao <fakzhao@cisco.com>
2025-02-28 08:50:22 -05:00
Daniel Rauber
186cd7f1a1
community: PlaywrightURLLoader should wait for page load event before attempting to extract data (#30043)
## Description

The PlaywrightURLLoader should wait for a page to be loaded before
attempting to extract data.
2025-02-28 08:45:51 -05:00
ccurme
0dbcc1d099
docs: document anthropic features (#30030)
Update integrations page with extended thinking feature.

Update API reference with extended thinking and citations.
2025-02-27 19:37:04 -05:00
ccurme
6c7c8a164f
openai[patch]: add unit test (#30022)
Test `max_completion_tokens` is propagated to payload for
AzureChatOpenAI.
2025-02-27 11:09:17 -05:00
DamonXue
156a60013a
docs: fix tavily_search code-block format. (#30012)
This pull request includes a change to the `TavilySearchResults` class
in the `tool.py` file, which updates the code block format in the
documentation.

Documentation update:

*
[`libs/community/langchain_community/tools/tavily_search/tool.py`](diffhunk://#diff-e3b6a980979268b639c6a86e9b182756b0f7c7e9e5605e613bc0a72ea6aa5301L54-R59):
Changed the code block format from Python to JSON in the example
provided in the docstring.Thank you for contributing to LangChain!
2025-02-27 10:55:15 -05:00
kawamou
8977ac5ab0
community[fix]: Handle None value in raw_content from Tavily API response (#30021)
## **Description:**

When using the Tavily retriever with include_raw_content=True, the
retriever occasionally fails with a Pydantic ValidationError because
raw_content can be None.

The Document model in langchain_core/documents/base.py requires
page_content to be a non-None value, but the Tavily API sometimes
returns None for raw_content.

This PR fixes the issue by ensuring that even when raw_content is None,
an empty string is used instead:

```python
page_content=result.get("content", "")
            if not self.include_raw_content
            else (result.get("raw_content") or ""),
2025-02-27 10:53:53 -05:00