108 Commits (5fcbe9dd2a789075c0f58828af385bca140fbba4)

Author	SHA1	Message	Date
Anis ZAKARI	37e89ba5b1	community[patch]: Bedrock add support for mistral models (#18756) ... *Description**: My previous [PR](https://github.com/langchain-ai/langchain/pull/18521) was mistakenly closed, so I am reopening this one. Context: AWS released two Mistral models on Bedrock last Friday (March 1, 2024). This PR includes some code adjustments to ensure their compatibility with the Bedrock class. --------- Co-authored-by: Anis ZAKARI <anis.zakari@hymaia.com> Co-authored-by: Erick Friis <erick@langchain.dev>	7 months ago
Phat Vo	3ecb903d49	community[patch] : Tidy up and update Clarifai SDK functions (#18314) ... Description : * Tidy up, add missing docstring and fix unused params * Enable using session token	7 months ago
Yunmo Koo	fee6f983ef	community[minor]: Integration for `Friendli` LLM and `ChatFriendli` ChatModel. (#17913) ... ## Description - Add [Friendli](https://friendli.ai/) integration for `Friendli` LLM and `ChatFriendli` chat model. - Unit tests and integration tests corresponding to this change are added. - Documentations corresponding to this change are added. ## Dependencies - Optional dependency [`friendli-client`](https://pypi.org/project/friendli-client/) package is added only for those who use `Frienldi` or `ChatFriendli` model. ## Twitter handle - https://twitter.com/friendliai	7 months ago
Erick Friis	1beb84b061	community[patch]: move pdf text tests to integration (#18746)	7 months ago
Guangdong Liu	ced5e7bae7	community[patch]: Fix sparkllm authentication problem. (#18651) ... - **Description:** fix sparkllm authentication problem.The current timestamp is in RFC1123 format. The time deviation must be controlled within 300s. I changed to re-obtain the url every time I ask a question. https://www.xfyun.cn/doc/spark/general_url_authentication.html#_1-2-%E9%89%B4%E6%9D%83%E5%8F%82%E6%95%B0	7 months ago
Piyush Jain	2b234a4d96	Support for claude v3 models. (#18630) ... Fixes #18513. ## Description This PR attempts to fix the support for Anthropic Claude v3 models in BedrockChat LLM. The changes here has updated the payload to use the `messages` format instead of the formatted text prompt for all models; `messages` API is backwards compatible with all models in Anthropic, so this should not break the experience for any models. ## Notes The PR in the current form does not support the v3 models for the non-chat Bedrock LLM. This means, that with these changes, users won't be able to able to use the v3 models with the Bedrock LLM. I can open a separate PR to tackle this use-case, the intent here was to get this out quickly, so users can start using and test the chat LLM. The Bedrock LLM classes have also grown complex with a lot of conditions to support various providers and models, and is ripe for a refactor to make future changes more palatable. This refactor is likely to take longer, and requires more thorough testing from the community. Credit to PRs [18579](https://github.com/langchain-ai/langchain/pull/18579) and [18548](https://github.com/langchain-ai/langchain/pull/18548) for some of the code here. --------- Co-authored-by: Erick Friis <erick@langchain.dev>	7 months ago
Eugene Yurtsev	4c25b49229	community[major]: breaking change in some APIs to force users to opt-in for pickling (#18696) ... This is a PR that adds a dangerous load parameter to force users to opt in to use pickle. This is a PR that's meant to raise user awareness that the pickling module is involved.	7 months ago
Liang Zhang	81985b31e6	community[patch]: Databricks SerDe uses cloudpickle instead of pickle (#18607) ... - **Description:** Databricks SerDe uses cloudpickle instead of pickle when serializing a user-defined function transform_input_fn since pickle does not support functions defined in `__main__`, and cloudpickle supports this. - **Dependencies:** cloudpickle>=2.0.0 Added a unit test.	7 months ago
Yudhajit Sinha	4570b477b9	community[patch]: Invoke callback prior to yielding token (titan_takeoff) (#18560) ... ## PR title community[patch]: Invoke callback prior to yielding token ## PR message - Description: Invoke callback prior to yielding token in _stream_ method in llms/titan_takeoff. - Issue: #16913 - Dependencies: None	7 months ago
Erick Friis	343438e872	community[patch]: deprecate community fireworks (#18544)	7 months ago
William De Vena	275877980e	community[patch]: Invoke callback prior to yielding token (#18447) ... ## PR title community[patch]: Invoke callback prior to yielding token ## PR message Description: Invoke callback prior to yielding token in _stream method in llms/vertexai. Issue: https://github.com/langchain-ai/langchain/issues/16913 Dependencies: None	7 months ago
William De Vena	67375e96e0	community[patch]: Invoke callback prior to yielding token (#18448) ... ## PR title community[patch]: Invoke callback prior to yielding token ## PR message - Description: Invoke callback prior to yielding token in _stream method in llms/tongyi. - Issue: https://github.com/langchain-ai/langchain/issues/16913 - Dependencies: None	7 months ago
William De Vena	eb04d0d3e2	community[patch]: Invoke callback prior to yielding token (#18452) ... ## PR title community[patch]: Invoke callback prior to yielding token ## PR message - Description: Invoke callback prior to yielding token in _stream and _astream methods in llms/anthropic. - Issue: https://github.com/langchain-ai/langchain/issues/16913 - Dependencies: None	7 months ago
William De Vena	371bec79bc	community[patch]: Invoke callback prior to yielding token (#18454) ... ## PR title community[patch]: Invoke callback prior to yielding token ## PR message - Description: Invoke callback prior to yielding token in _stream and _astream methods in llms/baidu_qianfan_endpoint. - Issue: https://github.com/langchain-ai/langchain/issues/16913 - Dependencies: None	7 months ago
Kate Silverstein	b7c71e2e07	community[minor]: llamafile embeddings support (#17976) ... * **Description:** adds `LlamafileEmbeddings` class implementation for generating embeddings using [llamafile](https://github.com/Mozilla-Ocho/llamafile)-based models. Includes related unit tests and notebook showing example usage. * **Issue:** N/A * **Dependencies:** N/A	7 months ago
Arun Sathiya	4adac20d7b	community[patch]: Make cohere_api_key a SecretStr (#12188) ... This PR makes `cohere_api_key` in `llms/cohere` a SecretStr, so that the API Key is not leaked when `Cohere.cohere_api_key` is represented as a string. --------- Signed-off-by: Arun <arun@arun.blog> Co-authored-by: Eugene Yurtsev <eyurtsev@gmail.com>	7 months ago
Nikita Titov	9f2ab37162	community[patch]: don't try to parse json in case of errored response (#18317) ... Related issue: #13896. In case Ollama is behind a proxy, proxy error responses cannot be viewed. You aren't even able to check response code. For example, if your Ollama has basic access authentication and it's not passed, `JSONDecodeError` will overwrite the truth response error. <details> <summary><b>Log now:</b></summary> ``` { "name": "JSONDecodeError", "message": "Expecting value: line 1 column 1 (char 0)", "stack": "--------------------------------------------------------------------------- JSONDecodeError Traceback (most recent call last) File /opt/miniforge3/envs/.gpt/lib/python3.10/site-packages/requests/models.py:971, in Response.json(self, **kwargs) 970 try: --> 971 return complexjson.loads(self.text, **kwargs) 972 except JSONDecodeError as e: 973 # Catch JSON-related errors and raise as requests.JSONDecodeError 974 # This aliases json.JSONDecodeError and simplejson.JSONDecodeError File /opt/miniforge3/envs/.gpt/lib/python3.10/json/__init__.py:346, in loads(s, cls, object_hook, parse_float, parse_int, parse_constant, object_pairs_hook, **kw) 343 if (cls is None and object_hook is None and 344 parse_int is None and parse_float is None and 345 parse_constant is None and object_pairs_hook is None and not kw): --> 346 return _default_decoder.decode(s) 347 if cls is None: File /opt/miniforge3/envs/.gpt/lib/python3.10/json/decoder.py:337, in JSONDecoder.decode(self, s, _w) 333 \"\"\"Return the Python representation of ``s`` (a ``str`` instance 334 containing a JSON document). 335 336 \"\"\" --> 337 obj, end = self.raw_decode(s, idx=_w(s, 0).end()) 338 end = _w(s, end).end() File /opt/miniforge3/envs/.gpt/lib/python3.10/json/decoder.py:355, in JSONDecoder.raw_decode(self, s, idx) 354 except StopIteration as err: --> 355 raise JSONDecodeError(\"Expecting value\", s, err.value) from None 356 return obj, end JSONDecodeError: Expecting value: line 1 column 1 (char 0) During handling of the above exception, another exception occurred: JSONDecodeError Traceback (most recent call last) Cell In[3], line 1 ----> 1 print(translate_func().invoke('text')) File /opt/miniforge3/envs/.gpt/lib/python3.10/site-packages/langchain_core/runnables/base.py:2053, in RunnableSequence.invoke(self, input, config) 2051 try: 2052 for i, step in enumerate(self.steps): -> 2053 input = step.invoke( 2054 input, 2055 # mark each step as a child run 2056 patch_config( 2057 config, callbacks=run_manager.get_child(f\"seq:step:{i+1}\") 2058 ), 2059 ) 2060 # finish the root run 2061 except BaseException as e: File /opt/miniforge3/envs/.gpt/lib/python3.10/site-packages/langchain_core/language_models/chat_models.py:165, in BaseChatModel.invoke(self, input, config, stop, **kwargs) 154 def invoke( 155 self, 156 input: LanguageModelInput, (...) 160 **kwargs: Any, 161 ) -> BaseMessage: 162 config = ensure_config(config) 163 return cast( 164 ChatGeneration, --> 165 self.generate_prompt( 166 [self._convert_input(input)], 167 stop=stop, 168 callbacks=config.get(\"callbacks\"), 169 tags=config.get(\"tags\"), 170 metadata=config.get(\"metadata\"), 171 run_name=config.get(\"run_name\"), 172 **kwargs, 173 ).generations[0][0], 174 ).message File /opt/miniforge3/envs/.gpt/lib/python3.10/site-packages/langchain_core/language_models/chat_models.py:543, in BaseChatModel.generate_prompt(self, prompts, stop, callbacks, **kwargs) 535 def generate_prompt( 536 self, 537 prompts: List[PromptValue], (...) 540 **kwargs: Any, 541 ) -> LLMResult: 542 prompt_messages = [p.to_messages() for p in prompts] --> 543 return self.generate(prompt_messages, stop=stop, callbacks=callbacks, **kwargs) File /opt/miniforge3/envs/.gpt/lib/python3.10/site-packages/langchain_core/language_models/chat_models.py:407, in BaseChatModel.generate(self, messages, stop, callbacks, tags, metadata, run_name, **kwargs) 405 if run_managers: 406 run_managers[i].on_llm_error(e, response=LLMResult(generations=[])) --> 407 raise e 408 flattened_outputs = [ 409 LLMResult(generations=[res.generations], llm_output=res.llm_output) 410 for res in results 411 ] 412 llm_output = self._combine_llm_outputs([res.llm_output for res in results]) File /opt/miniforge3/envs/.gpt/lib/python3.10/site-packages/langchain_core/language_models/chat_models.py:397, in BaseChatModel.generate(self, messages, stop, callbacks, tags, metadata, run_name, **kwargs) 394 for i, m in enumerate(messages): 395 try: 396 results.append( --> 397 self._generate_with_cache( 398 m, 399 stop=stop, 400 run_manager=run_managers[i] if run_managers else None, 401 **kwargs, 402 ) 403 ) 404 except BaseException as e: 405 if run_managers: File /opt/miniforge3/envs/.gpt/lib/python3.10/site-packages/langchain_core/language_models/chat_models.py:576, in BaseChatModel._generate_with_cache(self, messages, stop, run_manager, **kwargs) 572 raise ValueError( 573 \"Asked to cache, but no cache found at `langchain.cache`.\" 574 ) 575 if new_arg_supported: --> 576 return self._generate( 577 messages, stop=stop, run_manager=run_manager, **kwargs 578 ) 579 else: 580 return self._generate(messages, stop=stop, **kwargs) File /opt/miniforge3/envs/.gpt/lib/python3.10/site-packages/langchain_community/chat_models/ollama.py:250, in ChatOllama._generate(self, messages, stop, run_manager, **kwargs) 226 def _generate( 227 self, 228 messages: List[BaseMessage], (...) 231 **kwargs: Any, 232 ) -> ChatResult: 233 \"\"\"Call out to Ollama's generate endpoint. 234 235 Args: (...) 247 ]) 248 \"\"\" --> 250 final_chunk = self._chat_stream_with_aggregation( 251 messages, 252 stop=stop, 253 run_manager=run_manager, 254 verbose=self.verbose, 255 **kwargs, 256 ) 257 chat_generation = ChatGeneration( 258 message=AIMessage(content=final_chunk.text), 259 generation_info=final_chunk.generation_info, 260 ) 261 return ChatResult(generations=[chat_generation]) File /opt/miniforge3/envs/.gpt/lib/python3.10/site-packages/langchain_community/chat_models/ollama.py:183, in ChatOllama._chat_stream_with_aggregation(self, messages, stop, run_manager, verbose, **kwargs) 174 def _chat_stream_with_aggregation( 175 self, 176 messages: List[BaseMessage], (...) 180 **kwargs: Any, 181 ) -> ChatGenerationChunk: 182 final_chunk: Optional[ChatGenerationChunk] = None --> 183 for stream_resp in self._create_chat_stream(messages, stop, **kwargs): 184 if stream_resp: 185 chunk = _chat_stream_response_to_chat_generation_chunk(stream_resp) File /opt/miniforge3/envs/.gpt/lib/python3.10/site-packages/langchain_community/chat_models/ollama.py:156, in ChatOllama._create_chat_stream(self, messages, stop, **kwargs) 147 def _create_chat_stream( 148 self, 149 messages: List[BaseMessage], 150 stop: Optional[List[str]] = None, 151 **kwargs: Any, 152 ) -> Iterator[str]: 153 payload = { 154 \"messages\": self._convert_messages_to_ollama_messages(messages), 155 } --> 156 yield from self._create_stream( 157 payload=payload, stop=stop, api_url=f\"{self.base_url}/api/chat/\", **kwargs 158 ) File /opt/miniforge3/envs/.gpt/lib/python3.10/site-packages/langchain_community/llms/ollama.py:234, in _OllamaCommon._create_stream(self, api_url, payload, stop, **kwargs) 228 raise OllamaEndpointNotFoundError( 229 \"Ollama call failed with status code 404. \" 230 \"Maybe your model is not found \" 231 f\"and you should pull the model with `ollama pull {self.model}`.\" 232 ) 233 else: --> 234 optional_detail = response.json().get(\"error\") 235 raise ValueError( 236 f\"Ollama call failed with status code {response.status_code}.\" 237 f\" Details: {optional_detail}\" 238 ) 239 return response.iter_lines(decode_unicode=True) File /opt/miniforge3/envs/.gpt/lib/python3.10/site-packages/requests/models.py:975, in Response.json(self, **kwargs) 971 return complexjson.loads(self.text, **kwargs) 972 except JSONDecodeError as e: 973 # Catch JSON-related errors and raise as requests.JSONDecodeError 974 # This aliases json.JSONDecodeError and simplejson.JSONDecodeError --> 975 raise RequestsJSONDecodeError(e.msg, e.doc, e.pos) JSONDecodeError: Expecting value: line 1 column 1 (char 0)" } ``` </details> <details> <summary><b>Log after a fix:</b></summary> ``` { "name": "ValueError", "message": "Ollama call failed with status code 401. Details: <html>\r <head><title>401 Authorization Required</title></head>\r <body>\r <center><h1>401 Authorization Required</h1></center>\r <hr><center>nginx/1.18.0 (Ubuntu)</center>\r </body>\r </html>\r ", "stack": "--------------------------------------------------------------------------- ValueError Traceback (most recent call last) Cell In[2], line 1 ----> 1 print(translate_func().invoke('text')) File /opt/miniforge3/envs/.gpt/lib/python3.10/site-packages/langchain_core/runnables/base.py:2053, in RunnableSequence.invoke(self, input, config) 2051 try: 2052 for i, step in enumerate(self.steps): -> 2053 input = step.invoke( 2054 input, 2055 # mark each step as a child run 2056 patch_config( 2057 config, callbacks=run_manager.get_child(f\"seq:step:{i+1}\") 2058 ), 2059 ) 2060 # finish the root run 2061 except BaseException as e: File /opt/miniforge3/envs/.gpt/lib/python3.10/site-packages/langchain_core/language_models/chat_models.py:165, in BaseChatModel.invoke(self, input, config, stop, **kwargs) 154 def invoke( 155 self, 156 input: LanguageModelInput, (...) 160 **kwargs: Any, 161 ) -> BaseMessage: 162 config = ensure_config(config) 163 return cast( 164 ChatGeneration, --> 165 self.generate_prompt( 166 [self._convert_input(input)], 167 stop=stop, 168 callbacks=config.get(\"callbacks\"), 169 tags=config.get(\"tags\"), 170 metadata=config.get(\"metadata\"), 171 run_name=config.get(\"run_name\"), 172 **kwargs, 173 ).generations[0][0], 174 ).message File /opt/miniforge3/envs/.gpt/lib/python3.10/site-packages/langchain_core/language_models/chat_models.py:543, in BaseChatModel.generate_prompt(self, prompts, stop, callbacks, **kwargs) 535 def generate_prompt( 536 self, 537 prompts: List[PromptValue], (...) 540 **kwargs: Any, 541 ) -> LLMResult: 542 prompt_messages = [p.to_messages() for p in prompts] --> 543 return self.generate(prompt_messages, stop=stop, callbacks=callbacks, **kwargs) File /opt/miniforge3/envs/.gpt/lib/python3.10/site-packages/langchain_core/language_models/chat_models.py:407, in BaseChatModel.generate(self, messages, stop, callbacks, tags, metadata, run_name, **kwargs) 405 if run_managers: 406 run_managers[i].on_llm_error(e, response=LLMResult(generations=[])) --> 407 raise e 408 flattened_outputs = [ 409 LLMResult(generations=[res.generations], llm_output=res.llm_output) 410 for res in results 411 ] 412 llm_output = self._combine_llm_outputs([res.llm_output for res in results]) File /opt/miniforge3/envs/.gpt/lib/python3.10/site-packages/langchain_core/language_models/chat_models.py:397, in BaseChatModel.generate(self, messages, stop, callbacks, tags, metadata, run_name, **kwargs) 394 for i, m in enumerate(messages): 395 try: 396 results.append( --> 397 self._generate_with_cache( 398 m, 399 stop=stop, 400 run_manager=run_managers[i] if run_managers else None, 401 **kwargs, 402 ) 403 ) 404 except BaseException as e: 405 if run_managers: File /opt/miniforge3/envs/.gpt/lib/python3.10/site-packages/langchain_core/language_models/chat_models.py:576, in BaseChatModel._generate_with_cache(self, messages, stop, run_manager, **kwargs) 572 raise ValueError( 573 \"Asked to cache, but no cache found at `langchain.cache`.\" 574 ) 575 if new_arg_supported: --> 576 return self._generate( 577 messages, stop=stop, run_manager=run_manager, **kwargs 578 ) 579 else: 580 return self._generate(messages, stop=stop, **kwargs) File /opt/miniforge3/envs/.gpt/lib/python3.10/site-packages/langchain_community/chat_models/ollama.py:250, in ChatOllama._generate(self, messages, stop, run_manager, **kwargs) 226 def _generate( 227 self, 228 messages: List[BaseMessage], (...) 231 **kwargs: Any, 232 ) -> ChatResult: 233 \"\"\"Call out to Ollama's generate endpoint. 234 235 Args: (...) 247 ]) 248 \"\"\" --> 250 final_chunk = self._chat_stream_with_aggregation( 251 messages, 252 stop=stop, 253 run_manager=run_manager, 254 verbose=self.verbose, 255 **kwargs, 256 ) 257 chat_generation = ChatGeneration( 258 message=AIMessage(content=final_chunk.text), 259 generation_info=final_chunk.generation_info, 260 ) 261 return ChatResult(generations=[chat_generation]) File /storage/gpt-project/Repos/repo_nikita/gpt_lib/langchain/ollama.py:328, in ChatOllamaCustom._chat_stream_with_aggregation(self, messages, stop, run_manager, verbose, **kwargs) 319 def _chat_stream_with_aggregation( 320 self, 321 messages: List[BaseMessage], (...) 325 **kwargs: Any, 326 ) -> ChatGenerationChunk: 327 final_chunk: Optional[ChatGenerationChunk] = None --> 328 for stream_resp in self._create_chat_stream(messages, stop, **kwargs): 329 if stream_resp: 330 chunk = _chat_stream_response_to_chat_generation_chunk(stream_resp) File /storage/gpt-project/Repos/repo_nikita/gpt_lib/langchain/ollama.py:301, in ChatOllamaCustom._create_chat_stream(self, messages, stop, **kwargs) 292 def _create_chat_stream( 293 self, 294 messages: List[BaseMessage], 295 stop: Optional[List[str]] = None, 296 **kwargs: Any, 297 ) -> Iterator[str]: 298 payload = { 299 \"messages\": self._convert_messages_to_ollama_messages(messages), 300 } --> 301 yield from self._create_stream( 302 payload=payload, stop=stop, api_url=f\"{self.base_url}/api/chat\", **kwargs 303 ) File /storage/gpt-project/Repos/repo_nikita/gpt_lib/langchain/ollama.py:134, in _OllamaCommonCustom._create_stream(self, api_url, payload, stop, **kwargs) 132 else: 133 optional_detail = response.text --> 134 raise ValueError( 135 f\"Ollama call failed with status code {response.status_code}.\" 136 f\" Details: {optional_detail}\" 137 ) 138 return response.iter_lines(decode_unicode=True) ValueError: Ollama call failed with status code 401. Details: <html>\r <head><title>401 Authorization Required</title></head>\r <body>\r <center><h1>401 Authorization Required</h1></center>\r <hr><center>nginx/1.18.0 (Ubuntu)</center>\r </body>\r </html>\r " } ``` </details> The same is true for timeout errors or when you simply mistyped in `base_url` arg and get response from some other service, for instance. Real Ollama errors are still clearly readable: ``` ValueError: Ollama call failed with status code 400. Details: {"error":"invalid options: unknown_option"} ``` --------- Co-authored-by: Bagatur <baskaryan@gmail.com>	7 months ago
Guangdong Liu	760a16ff32	community[patch]: Fix ChatModel for sparkllm Bug. (#18375) ... **PR message**: ***Delete this entire checklist*** and replace with - **Description:** fix sparkllm paramer error - **Issue:** close #18370 - **Dependencies:** change `IFLYTEK_SPARK_APP_URL` to `IFLYTEK_SPARK_API_URL` - **Twitter handle:** No	7 months ago
Shengsheng Huang	ae471a7dcb	community[minor]: add BigDL-LLM integrations (#17953) ... - **Description**: [`bigdl-llm`](https://github.com/intel-analytics/BigDL) is a library for running LLM on Intel XPU (from Laptop to GPU to Cloud) using INT4/FP4/INT8/FP8 with very low latency (for any PyTorch model). This PR adds bigdl-llm integrations to langchain. - **Issue**: NA - **Dependencies**: `bigdl-llm` library - **Contribution maintainer**: @shane-huang Examples added: - docs/docs/integrations/llms/bigdl.ipynb	7 months ago
Ethan Yang	f61cb8d407	community[minor]: Add openvino backend support (#11591) ... - **Description:** add openvino backend support by HuggingFace Optimum Intel, - **Dependencies:** “optimum[openvino]”, --------- Co-authored-by: Bagatur <baskaryan@gmail.com>	7 months ago
Erick Friis	eefb49680f	multiple[patch]: fix deprecation versions (#18349)	7 months ago
William De Vena	6b58943917	community[patch]: Invoke callback prior to yielding token (#18288) ... ## PR title community[patch]: Invoke callback prior to yielding PR message Description: Invoke on_llm_new_token callback prior to yielding token in _stream and _astream methods. Issue: https://github.com/langchain-ai/langchain/issues/16913 Dependencies: None Twitter handle: None	7 months ago
kYLe	17ecf6e119	community[patch]: Remove model limitation on Anyscale LLM (#17662) ... **Description:** Llama Guard is deprecated from Anyscale public endpoint. **Issue:** Change the default model. and remove the limitation of only use Llama Guard with Anyscale LLMs Anyscale LLM can also works with all other Chat model hosted on Anyscale. Also added `async_client` for Anyscale LLM	7 months ago
Erick Friis	29e0445490	community[patch]: BaseLLM typing in init (#18029)	7 months ago
Guangdong Liu	4197efd67a	community: Fix SparkLLM error (#18015) ... Thank you for contributing to LangChain! - [x] **PR title**: "package: description" - Where "package" is whichever of langchain, community, core, experimental, etc. is being modified. Use "docs: ..." for purely docs changes, "templates: ..." for template changes, "infra: ..." for CI changes. - Example: "community: add foobar LLM" - **Description:** fix SparkLLM error - **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!	7 months ago
Brad Erickson	ecd72d26cf	community: Bugfix - correct Ollama API path to avoid HTTP 307 (#17895) ... Sets the correct /api/generate path, without ending /, to reduce HTTP requests. Reference: https://github.com/ollama/ollama/blob/efe040f8/docs/api.md#generate-request-streaming Before: DEBUG: Starting new HTTP connection (1): localhost:11434 DEBUG: http://localhost:11434 "POST /api/generate/ HTTP/1.1" 307 0 DEBUG: http://localhost:11434 "POST /api/generate HTTP/1.1" 200 None After: DEBUG: Starting new HTTP connection (1): localhost:11434 DEBUG: http://localhost:11434 "POST /api/generate HTTP/1.1" 200 None	7 months ago
Guangdong Liu	47b1b7092d	community[minor]: Add SparkLLM to community (#17702)	7 months ago
Aymeric Roucher	0d294760e7	Community: Fuse HuggingFace Endpoint-related classes into one (#17254) ... ## Description Fuse HuggingFace Endpoint-related classes into one: - [HuggingFaceHub](5ceaf784f3/libs/community/langchain_community/llms/huggingface_hub.py) - [HuggingFaceTextGenInference](5ceaf784f3/libs/community/langchain_community/llms/huggingface_text_gen_inference.py) - and [HuggingFaceEndpoint](5ceaf784f3/libs/community/langchain_community/llms/huggingface_endpoint.py) Are fused into - HuggingFaceEndpoint ## Issue The deduplication of classes was creating a lack of clarity, and additional effort to develop classes leads to issues like [this hack](5ceaf784f3/libs/community/langchain_community/llms/huggingface_endpoint.py (L159)). ## Dependancies None, this removes dependancies. ## Twitter handle If you want to post about this: @AymericRoucher --------- Co-authored-by: Bagatur <baskaryan@gmail.com>	7 months ago
Mohammad Mohtashim	43dc5d3416	community[patch]: OpenLLM Client Fixes + Added Timeout Parameter (#17478) ... - OpenLLM was using outdated method to get the final text output from openllm client invocation which was raising the error. Therefore corrected that. - OpenLLM `_identifying_params` was getting the openllm's client configuration using outdated attributes which was raising error. - Updated the docstring for OpenLLM. - Added timeout parameter to be passed to underlying openllm client.	7 months ago
Mateusz Szewczyk	916332ef5b	ibm: added partners package `langchain_ibm`, added llm (#16512) ... - **Description:** Added `langchain_ibm` as an langchain partners package of IBM [watsonx.ai](https://www.ibm.com/products/watsonx-ai) LLM provider (`WatsonxLLM`) - **Dependencies:** [ibm-watsonx-ai](https://pypi.org/project/ibm-watsonx-ai/), - **Tag maintainer:** : --------- Co-authored-by: Erick Friis <erick@langchain.dev>	7 months ago
wulixuan	c776cfc599	community[minor]: integrate with model Yuan2.0 (#15411) ... 1. integrate with [`Yuan2.0`](https://github.com/IEIT-Yuan/Yuan-2.0/blob/main/README-EN.md) 2. update `langchain.llms` 3. add a new doc for [Yuan2.0 integration](docs/docs/integrations/llms/yuan2.ipynb) --------- Co-authored-by: Harrison Chase <hw.chase.17@gmail.com> Co-authored-by: Bagatur <baskaryan@gmail.com>	7 months ago
Alex Peplowski	70c296ae96	community[patch]: Expose Anthropic Retry Logic (#17069) ... **Description:** Expose Anthropic's retry logic, so that `max_retries` can be configured via langchain. Anthropic's retry logic is implemented in their Python SDK here: https://github.com/anthropics/anthropic-sdk-python?tab=readme-ov-file#retries --------- Co-authored-by: Bagatur <baskaryan@gmail.com>	7 months ago
Kate Silverstein	0bc4a9b3fc	community[minor]: Adds Llamafile as an LLM (#17431) ... * **Description:** Adds a simple LLM implementation for interacting with [llamafile](https://github.com/Mozilla-Ocho/llamafile)-based models. * **Dependencies:** N/A * **Issue:** N/A **Detail** [llamafile](https://github.com/Mozilla-Ocho/llamafile) lets you run LLMs locally from a single file on most computers without installing any dependencies. To use the llamafile LLM implementation, the user needs to: 1. Download a llamafile e.g. https://huggingface.co/jartine/TinyLlama-1.1B-Chat-v1.0-GGUF/resolve/main/TinyLlama-1.1B-Chat-v1.0.Q5_K_M.llamafile?download=true 2. Make the file executable. 3. Run the llamafile in 'server mode'. (All llamafiles come packaged with a lightweight server; by default, the server listens at `http://localhost:8080`.) ```bash wget https://url/of/model.llamafile chmod +x model.llamafile ./model.llamafile --server --nobrowser ``` Now, the user can invoke the LLM via the LangChain client: ```python from langchain_community.llms.llamafile import Llamafile llm = Llamafile() llm.invoke("Tell me a joke.") ```	7 months ago
Nat Noordanus	8a3b74fe1f	community[patch]: Fix pydantic ForwardRef error in BedrockBase (#17416) ... - **Description:** Fixes a type annotation issue in the definition of BedrockBase. This issue was that the annotation for the `config` attribute includes a ForwardRef to `botocore.client.Config` which is only imported when `TYPE_CHECKING`. This can cause pydantic to raise an error like `pydantic.errors.ConfigError: field "config" not yet prepared so type is still a ForwardRef, ...`. - **Issue:** N/A - **Dependencies:** N/A - **Twitter handle:** `@__nat_n__`	7 months ago
Theo / Taeyoon Kang	1987f905ed	core[patch]: Support .yml extension for YAML (#16783) ... - **Description:** [AS-IS] When dealing with a yaml file, the extension must be .yaml. [TO-BE] In the absence of extension length constraints in the OS, the extension of the YAML file is yaml, but control over the yml extension must still be made. It's as if it's an error because it's a .jpg extension in jpeg support. - **Issue:** - - **Dependencies:** no dependencies required for this change,	7 months ago
Robby	0653aa469a	community[patch]: Invoke callback prior to yielding token (#17346) ... **Description:** Invoke callback prior to yielding token in stream method for watsonx. **Issue:** [Callback for on_llm_new_token should be invoked before the token is yielded by the model #16913](https://github.com/langchain-ai/langchain/issues/16913) Co-authored-by: Robby <h0rv@users.noreply.github.com>	7 months ago
Erick Friis	3a2eb6e12b	infra: add print rule to ruff (#16221) ... Added noqa for existing prints. Can slowly remove / will prevent more being intro'd	8 months ago
kYLe	c9999557bf	community[patch]: Modify LLMs/Anyscale work with OpenAI API v1 (#14206) ... <!-- 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. --> - **Description:** 1. Modify LLMs/Anyscale to work with OAI v1 2. Get rid of openai_ prefixed variables in Chat_model/ChatAnyscale 3. Modify `anyscale_api_base` to `anyscale_base_url` to follow OAI name convention (reverted) --------- Co-authored-by: Erick Friis <erick@langchain.dev>	8 months ago
Leonid Ganeline	932c52c333	community[patch]: docstrings (#16810) ... - added missed docstrings - formated docstrings to the consistent form	8 months ago
Armin Stepanyan	641efcf41c	community: add runtime kwargs to HuggingFacePipeline (#17005) ... This PR enables changing the behaviour of huggingface pipeline between different calls. For example, before this PR there's no way of changing maximum generation length between different invocations of the chain. This is desirable in cases, such as when we want to scale the maximum output size depending on a dynamic prompt size. Usage example: ```python from langchain_community.llms.huggingface_pipeline import HuggingFacePipeline from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline model_id = "gpt2" tokenizer = AutoTokenizer.from_pretrained(model_id) model = AutoModelForCausalLM.from_pretrained(model_id) pipe = pipeline("text-generation", model=model, tokenizer=tokenizer) hf = HuggingFacePipeline(pipeline=pipe) hf("Say foo:", pipeline_kwargs={"max_new_tokens": 42}) ``` --------- Co-authored-by: Bagatur <baskaryan@gmail.com>	8 months ago
Liang Zhang	7306600e2f	community[patch]: Support SerDe transform functions in Databricks LLM (#16752) ... **Description:** Databricks LLM does not support SerDe the transform_input_fn and transform_output_fn. After saving and loading, the LLM will be broken. This PR serialize these functions into a hex string using pickle, and saving the hex string in the yaml file. Using pickle to serialize a function can be flaky, but this is a simple workaround that unblocks many use cases. If more sophisticated SerDe is needed, we can improve it later. Test: Added a simple unit test. I did manual test on Databricks and it works well. The saved yaml looks like: ``` llm: _type: databricks cluster_driver_port: null cluster_id: null databricks_uri: databricks endpoint_name: databricks-mixtral-8x7b-instruct extra_params: {} host: e2-dogfood.staging.cloud.databricks.com max_tokens: null model_kwargs: null n: 1 stop: null task: null temperature: 0.0 transform_input_fn: 80049520000000000000008c085f5f6d61696e5f5f948c0f7472616e73666f726d5f696e7075749493942e transform_output_fn: null ``` @baskaryan ```python from langchain_community.embeddings import DatabricksEmbeddings from langchain_community.llms import Databricks from langchain.chains import RetrievalQA from langchain.document_loaders import TextLoader from langchain.text_splitter import CharacterTextSplitter from langchain.vectorstores import FAISS import mlflow embeddings = DatabricksEmbeddings(endpoint="databricks-bge-large-en") def transform_input(**request): request["messages"] = [ { "role": "user", "content": request["prompt"] } ] del request["prompt"] return request llm = Databricks(endpoint_name="databricks-mixtral-8x7b-instruct", transform_input_fn=transform_input) persist_dir = "faiss_databricks_embedding" # Create the vector db, persist the db to a local fs folder loader = TextLoader("state_of_the_union.txt") documents = loader.load() text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=0) docs = text_splitter.split_documents(documents) db = FAISS.from_documents(docs, embeddings) db.save_local(persist_dir) def load_retriever(persist_directory): embeddings = DatabricksEmbeddings(endpoint="databricks-bge-large-en") vectorstore = FAISS.load_local(persist_directory, embeddings) return vectorstore.as_retriever() retriever = load_retriever(persist_dir) retrievalQA = RetrievalQA.from_llm(llm=llm, retriever=retriever) with mlflow.start_run() as run: logged_model = mlflow.langchain.log_model( retrievalQA, artifact_path="retrieval_qa", loader_fn=load_retriever, persist_dir=persist_dir, ) # Load the retrievalQA chain loaded_model = mlflow.pyfunc.load_model(logged_model.model_uri) print(loaded_model.predict([{"query": "What did the president say about Ketanji Brown Jackson"}])) ```	8 months ago
François Paupier	929f071513	community[patch]: Fix error in `LlamaCpp` community LLM with Configurable Fields, 'grammar' custom type not available (#16995) ... - **Description:** Ensure the `LlamaGrammar` custom type is always available when instantiating a `LlamaCpp` LLM - **Issue:** #16994 - **Dependencies:** None - **Twitter handle:** @fpaupier --------- Co-authored-by: Bagatur <baskaryan@gmail.com>	8 months ago
Mohammad Mohtashim	3c4b24b69a	community[patch]: Fix the _call of HuggingFaceHub (#16891) ... Fixed the following identified issue: #16849 @baskaryan	8 months ago
Supreet Takkar	ae33979813	community[patch]: Allow adding ARNs as model_id to support Amazon Bedrock custom models (#16800) ... - **Description:** Adds an additional class variable to `BedrockBase` called `provider` that allows sending a model provider such as amazon, cohere, ai21, etc. Up until now, the model provider is extracted from the `model_id` using the first part before the `.`, such as `amazon` for `amazon.titan-text-express-v1` (see [supported list of Bedrock model IDs here](https://docs.aws.amazon.com/bedrock/latest/userguide/model-ids-arns.html)). But for custom Bedrock models where the ARN of the provisioned throughput must be supplied, the `model_id` is like `arn:aws:bedrock:...` so the `model_id` cannot be extracted from this. A model `provider` is required by the LangChain Bedrock class to perform model-based processing. To allow the same processing to be performed for custom-models of a specific base model type, passing this `provider` argument can help solve the issues. The alternative considered here was the use of `provider.arn:aws:bedrock:...` which then requires ARN to be extracted and passed separately when invoking the model. The proposed solution here is simpler and also does not cause issues for current models already using the Bedrock class. - **Issue:** N/A - **Dependencies:** N/A --------- Co-authored-by: Piyush Jain <piyushjain@duck.com>	8 months ago
Bagatur	66e45e8ab7	community[patch]: chat model mypy fixes (#17061) ... Related to #17048	8 months ago
Harrison Chase	4eda647fdd	infra: add -p to mkdir in lint steps (#17013) ... Previously, if this did not find a mypy cache then it wouldnt run this makes it always run adding mypy ignore comments with existing uncaught issues to unblock other prs --------- Co-authored-by: Erick Friis <erick@langchain.dev> Co-authored-by: Bagatur <22008038+baskaryan@users.noreply.github.com>	8 months ago
Bob Lin	546b757303	community: Add ChatGLM3 (#15265) ... Add [ChatGLM3](https://github.com/THUDM/ChatGLM3) and updated [chatglm.ipynb](https://python.langchain.com/docs/integrations/llms/chatglm) --------- Co-authored-by: Bagatur <baskaryan@gmail.com> Co-authored-by: Harrison Chase <hw.chase.17@gmail.com>	8 months ago
baichuan-assistant	f8f2649f12	community: Add Baichuan LLM to community (#16724) ... Replace this entire comment with: - **Description:** Add Baichuan LLM to integration/llm, also updated related docs. Co-authored-by: BaiChuanHelper <wintergyc@WinterGYCs-MacBook-Pro.local>	8 months ago
hulitaitai	32cad38ec6	<langchain_community\llms\chatglm.py>: <Correcting "history"> (#16729) ... Use the real "history" provided by the original program instead of putting "None" in the history. - **Description:** I change one line in the code to make it return the "history" of the chat model. - **Issue:** At the moment it returns only the answers of the chat model. However the chat model himself provides a history more complet with the questions of the user. - **Dependencies:** no dependencies required for this change,	8 months ago
Bassem Yacoube	85e93e05ed	community[minor]: Update OctoAI LLM, Embedding and documentation (#16710) ... This PR includes updates for OctoAI integrations: - The LLM class was updated to fix a bug that occurs with multiple sequential calls - The Embedding class was updated to support the new GTE-Large endpoint released on OctoAI lately - The documentation jupyter notebook was updated to reflect using the new LLM sdk Thank you!	8 months ago