github/langchain
1
0
Fork 0
mirror of https://github.com/hwchase17/langchain.git synced 2025-08-15 07:36:08 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'master' into pprados/06-pdfplumber

Browse Source
This commit is contained in:
Philippe PRADOS 2025-02-28 08:35:36 +01:00 committed by GitHub
commit 0cf0f70c45
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
11 changed files with 305 additions and 140 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

53
docs/docs/integrations/chat/anthropic.ipynb
View File

@ -315,6 +315,59 @@
"ai_msg.tool_calls" "ai_msg.tool_calls"
] ]
}, },
{
"cell_type": "markdown",
"id": "6e36d25c-f358-49e5-aefa-b99fbd3fec6b",
"metadata": {},
"source": [
"## Extended thinking\n",
"\n",
"Claude 3.7 Sonnet supports an [extended thinking](https://docs.anthropic.com/en/docs/build-with-claude/extended-thinking) feature, which will output the step-by-step reasoning process that led to its final answer.\n",
"\n",
"To use it, specify the `thinking` parameter when initializing `ChatAnthropic`. It can also be passed in as a kwarg during invocation.\n",
"\n",
"You will need to specify a token budget to use this feature. See usage example below:"
]
},
{
"cell_type": "code",
"execution_count": 1,
"id": "a34cf93b-8522-43a6-a3f3-8a189ddf54a7",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"[\n",
" {\n",
" \"signature\": \"ErUBCkYIARgCIkCx7bIPj35jGPHpoVOB2y5hvPF8MN4lVK75CYGftmVNlI4axz2+bBbSexofWsN1O/prwNv8yPXnIXQmwT6zrJsKEgwJzvks0yVRZtaGBScaDOm9xcpOxbuhku1zViIw9WDgil/KZL8DsqWrhVpC6TzM0RQNCcsHcmgmyxbgG9g8PR0eJGLxCcGoEw8zMQu1Kh1hQ1/03hZ2JCOgigpByR9aNPTwwpl64fQUe6WwIw==\",\n",
" \"thinking\": \"To find the cube root of 50.653, I need to find the value of $x$ such that $x^3 = 50.653$.\\n\\nI can try to estimate this first. \\n$3^3 = 27$\\n$4^3 = 64$\\n\\nSo the cube root of 50.653 will be somewhere between 3 and 4, but closer to 4.\\n\\nLet me try to compute this more precisely. I can use the cube root function:\\n\\ncube root of 50.653 = 50.653^(1/3)\\n\\nLet me calculate this:\\n50.653^(1/3) \\u2248 3.6998\\n\\nLet me verify:\\n3.6998^3 \\u2248 50.6533\\n\\nThat's very close to 50.653, so I'm confident that the cube root of 50.653 is approximately 3.6998.\\n\\nActually, let me compute this more precisely:\\n50.653^(1/3) \\u2248 3.69981\\n\\nLet me verify once more:\\n3.69981^3 \\u2248 50.652998\\n\\nThat's extremely close to 50.653, so I'll say that the cube root of 50.653 is approximately 3.69981.\",\n",
" \"type\": \"thinking\"\n",
" },\n",
" {\n",
" \"text\": \"The cube root of 50.653 is approximately 3.6998.\\n\\nTo verify: 3.6998\\u00b3 = 50.6530, which is very close to our original number.\",\n",
" \"type\": \"text\"\n",
" }\n",
"]\n"
]
}
],
"source": [
"import json\n",
"\n",
"from langchain_anthropic import ChatAnthropic\n",
"\n",
"llm = ChatAnthropic(\n",
" model=\"claude-3-7-sonnet-latest\",\n",
" max_tokens=5000,\n",
" thinking={\"type\": \"enabled\", \"budget_tokens\": 2000},\n",
")\n",
"\n",
"response = llm.invoke(\"What is the cube root of 50.653?\")\n",
"print(json.dumps(response.content, indent=2))"
]
},
{ {
"cell_type": "markdown", "cell_type": "markdown",
"id": "301d372f-4dec-43e6-b58c-eee25633e1a6", "id": "301d372f-4dec-43e6-b58c-eee25633e1a6",

94
docs/docs/integrations/chat/writer.ipynb
View File

@ -16,8 +16,8 @@
"\n", "\n",
"### Integration details\n", "### Integration details\n",
"| Class | Package | Local | Serializable | JS support | Package downloads | Package latest |\n", "| Class | Package | Local | Serializable | JS support | Package downloads | Package latest |\n",
"|:--------------------------------------------------------------------------------------------------------------------------------------------|:-----------------| :---: | :---: |:----------:|:------------------------------------------------------------------------------------------------:|:---------------------------------------------------------------------------------------------:|\n", "|:-------------------------------------------------------------------------------------------------------------------------|:-----------------| :---: | :---: |:----------:|:------------------------------------------------------------------------------------------------:|:---------------------------------------------------------------------------------------------:|\n",
"| ChatWriter | [langchain-writer](https://pypi.org/project/langchain-writer/) | ❌ | ❌ | ❌ | ![PyPI - Downloads](https://img.shields.io/pypi/dm/langchain-writer?style=flat-square&label=%20) | ![PyPI - Version](https://img.shields.io/pypi/v/langchain-writer?style=flat-square&label=%20) |\n", "| [ChatWriter](https://github.com/writer/langchain-writer/blob/main/langchain_writer/chat_models.py#L308) | [langchain-writer](https://pypi.org/project/langchain-writer/) | ❌ | ❌ | ❌ | ![PyPI - Downloads](https://img.shields.io/pypi/dm/langchain-writer?style=flat-square&label=%20) | ![PyPI - Version](https://img.shields.io/pypi/v/langchain-writer?style=flat-square&label=%20) |\n",
"### Model features\n", "### Model features\n",
"| [Tool calling](/docs/how_to/tool_calling) | Structured output | JSON mode | Image input | Audio input | Video input | [Token-level streaming](/docs/how_to/chat_streaming/) | Native async | [Token usage](/docs/how_to/chat_token_usage_tracking/) | Logprobs |\n", "| [Tool calling](/docs/how_to/tool_calling) | Structured output | JSON mode | Image input | Audio input | Video input | [Token-level streaming](/docs/how_to/chat_streaming/) | Native async | [Token usage](/docs/how_to/chat_token_usage_tracking/) | Logprobs |\n",
"| :---: |:-----------------:| :---: | :---: | :---: | :---: | :---: | :---: |:--------------------------------:|:--------:|\n", "| :---: |:-----------------:| :---: | :---: | :---: | :---: | :---: | :---: |:--------------------------------:|:--------:|\n",
@ -36,17 +36,21 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "433e8d2b-9519-4b49-b2c4-7ab65b046c94", "id": "433e8d2b-9519-4b49-b2c4-7ab65b046c94",
"metadata": {}, "metadata": {
"outputs": [], "jupyter": {
"is_executing": true
}
},
"source": [ "source": [
"import getpass\n", "import getpass\n",
"import os\n", "import os\n",
"\n", "\n",
"if not os.getenv(\"WRITER_API_KEY\"):\n", "if not os.getenv(\"WRITER_API_KEY\"):\n",
" os.environ[\"WRITER_API_KEY\"] = getpass.getpass(\"Enter your Writer API key: \")" " os.environ[\"WRITER_API_KEY\"] = getpass.getpass(\"Enter your Writer API key: \")"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -58,14 +62,14 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "a15d341e-3e26-4ca3-830b-5aab30ed66de", "id": "a15d341e-3e26-4ca3-830b-5aab30ed66de",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"# os.environ[\"LANGCHAIN_TRACING_V2\"] = \"true\"\n", "# os.environ[\"LANGCHAIN_TRACING_V2\"] = \"true\"\n",
"# os.environ[\"LANGCHAIN_API_KEY\"] = getpass.getpass(\"Enter your LangSmith API key: \")" "# os.environ[\"LANGCHAIN_API_KEY\"] = getpass.getpass(\"Enter your LangSmith API key: \")"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -79,13 +83,13 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "652d6238-1f87-422a-b135-f5abbb8652fc", "id": "652d6238-1f87-422a-b135-f5abbb8652fc",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"%pip install -qU langchain-writer" "%pip install -qU langchain-writer"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -99,10 +103,8 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "cb09c344-1836-4e0c-acf8-11d13ac1dbae", "id": "cb09c344-1836-4e0c-acf8-11d13ac1dbae",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"from langchain_writer import ChatWriter\n", "from langchain_writer import ChatWriter\n",
"\n", "\n",
@ -113,7 +115,9 @@
" timeout=None,\n", " timeout=None,\n",
" max_retries=2,\n", " max_retries=2,\n",
")" ")"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -127,12 +131,10 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "62e0dbc3", "id": "62e0dbc3",
"metadata": { "metadata": {
"tags": [] "tags": []
}, },
"outputs": [],
"source": [ "source": [
"messages = [\n", "messages = [\n",
" (\n", " (\n",
@ -143,7 +145,9 @@
"]\n", "]\n",
"ai_msg = llm.invoke(messages)\n", "ai_msg = llm.invoke(messages)\n",
"ai_msg" "ai_msg"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -155,13 +159,13 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "d86145b3-bfef-46e8-b227-4dda5c9c2705", "id": "d86145b3-bfef-46e8-b227-4dda5c9c2705",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"print(ai_msg.content)" "print(ai_msg.content)"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -175,10 +179,8 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "4a0f2112b3a4c79e", "id": "4a0f2112b3a4c79e",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"messages = [\n", "messages = [\n",
" (\n", " (\n",
@ -189,7 +191,9 @@
"]\n", "]\n",
"ai_stream = llm.stream(messages)\n", "ai_stream = llm.stream(messages)\n",
"ai_stream" "ai_stream"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -201,14 +205,14 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "8c4b7b9b9308c757", "id": "8c4b7b9b9308c757",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"for chunk in ai_stream:\n", "for chunk in ai_stream:\n",
" print(chunk.content, end=\"\")" " print(chunk.content, end=\"\")"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -236,10 +240,8 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "47e2f0faceca533", "id": "47e2f0faceca533",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"from pydantic import BaseModel, Field\n", "from pydantic import BaseModel, Field\n",
"\n", "\n",
@ -251,7 +253,9 @@
"\n", "\n",
"\n", "\n",
"llm.bind_tools([GetWeather])" "llm.bind_tools([GetWeather])"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -263,16 +267,16 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "765527dd533ec967", "id": "765527dd533ec967",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"ai_msg = llm.invoke(\n", "ai_msg = llm.invoke(\n",
" \"what is the weather like in New York City\",\n", " \"what is the weather like in New York City\",\n",
")\n", ")\n",
"ai_msg" "ai_msg"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -284,13 +288,13 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "f361c4769e772fe", "id": "f361c4769e772fe",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"print(ai_msg.tool_calls)" "print(ai_msg.tool_calls)"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -316,10 +320,8 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "c8a217f6190747fe", "id": "c8a217f6190747fe",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"ai_batch = llm.batch(\n", "ai_batch = llm.batch(\n",
" [\n", " [\n",
@ -330,7 +332,9 @@
" config={\"max_concurrency\": 3},\n", " config={\"max_concurrency\": 3},\n",
")\n", ")\n",
"ai_batch" "ai_batch"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -342,15 +346,15 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "b6a228d448f3df23", "id": "b6a228d448f3df23",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"for batch in ai_batch:\n", "for batch in ai_batch:\n",
" print(batch.content)\n", " print(batch.content)\n",
" print(\"-\" * 100)" " print(\"-\" * 100)"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -374,10 +378,8 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "e197d1d7-a070-4c96-9f8a-a0e86d046e0b", "id": "e197d1d7-a070-4c96-9f8a-a0e86d046e0b",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"from langchain_core.prompts import ChatPromptTemplate\n", "from langchain_core.prompts import ChatPromptTemplate\n",
"\n", "\n",
@ -399,7 +401,9 @@
" \"input\": \"I love programming.\",\n", " \"input\": \"I love programming.\",\n",
" }\n", " }\n",
")" ")"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",

48
docs/docs/integrations/document_loaders/parsers/writer_pdf_parser.ipynb
View File

@ -15,8 +15,8 @@
"\n", "\n",
"### Integration details\n", "### Integration details\n",
"| Class | Package | Local | Serializable | JS support | Package downloads | Package latest |\n", "| Class | Package | Local | Serializable | JS support | Package downloads | Package latest |\n",
"|:------------------------------------------------------------------------------------------------------------------------------------------|:-----------------| :---: | :---: |:----------:|:------------------------------------------------------------------------------------------------:|:---------------------------------------------------------------------------------------------:|\n", "|:-----------------------------------------------------------------------------------------------------------------------------------|:-----------------| :---: | :---: |:----------:|:------------------------------------------------------------------------------------------------:|:---------------------------------------------------------------------------------------------:|\n",
"| PDFParser | [langchain-writer](https://pypi.org/project/langchain-writer/) | ❌ | ❌ | ❌ | ![PyPI - Downloads](https://img.shields.io/pypi/dm/langchain-writer?style=flat-square&label=%20) | ![PyPI - Version](https://img.shields.io/pypi/v/langchain-writer?style=flat-square&label=%20) |" "| [PDFParser](https://github.com/writer/langchain-writer/blob/main/langchain_writer/pdf_parser.py#L55) | [langchain-writer](https://pypi.org/project/langchain-writer/) | ❌ | ❌ | ❌ | ![PyPI - Downloads](https://img.shields.io/pypi/dm/langchain-writer?style=flat-square&label=%20) | ![PyPI - Version](https://img.shields.io/pypi/v/langchain-writer?style=flat-square&label=%20) |"
] ]
}, },
{ {
@ -31,17 +31,13 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "a8d653f15b7ee32d", "id": "a8d653f15b7ee32d",
"metadata": { "metadata": {},
"jupyter": {
"is_executing": true
}
},
"outputs": [],
"source": [ "source": [
"%pip install --quiet -U langchain-writer" "%pip install --quiet -U langchain-writer"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -55,17 +51,17 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "2983e19c9d555e58", "id": "2983e19c9d555e58",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"import getpass\n", "import getpass\n",
"import os\n", "import os\n",
"\n", "\n",
"if not os.getenv(\"WRITER_API_KEY\"):\n", "if not os.getenv(\"WRITER_API_KEY\"):\n",
" os.environ[\"WRITER_API_KEY\"] = getpass.getpass(\"Enter your Writer API key: \")" " os.environ[\"WRITER_API_KEY\"] = getpass.getpass(\"Enter your Writer API key: \")"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -77,14 +73,14 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "98d8422ecee77403", "id": "98d8422ecee77403",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"# os.environ[\"LANGCHAIN_TRACING_V2\"] = \"true\"\n", "# os.environ[\"LANGCHAIN_TRACING_V2\"] = \"true\"\n",
"# os.environ[\"LANGCHAIN_API_KEY\"] = getpass.getpass()" "# os.environ[\"LANGCHAIN_API_KEY\"] = getpass.getpass()"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -98,15 +94,15 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "787b3ba8af32533f", "id": "787b3ba8af32533f",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"from langchain_writer.pdf_parser import PDFParser\n", "from langchain_writer.pdf_parser import PDFParser\n",
"\n", "\n",
"parser = PDFParser(format=\"markdown\")" "parser = PDFParser(format=\"markdown\")"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -124,18 +120,18 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "d1a24b81a8a96f09", "id": "d1a24b81a8a96f09",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"from langchain_core.documents.base import Blob\n", "from langchain_core.documents.base import Blob\n",
"\n", "\n",
"file = Blob.from_path(\"../../data/page_to_parse.pdf\")\n", "file = Blob.from_path(\"../example_data/layout-parser-paper.pdf\")\n",
"\n", "\n",
"parsed_pages = parser.parse(blob=file)\n", "parsed_pages = parser.parse(blob=file)\n",
"parsed_pages" "parsed_pages"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -149,14 +145,14 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "e2f7fd52b7188c6c", "id": "e2f7fd52b7188c6c",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"parsed_pages_async = await parser.aparse(blob=file)\n", "parsed_pages_async = await parser.aparse(blob=file)\n",
"parsed_pages_async" "parsed_pages_async"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",

2
docs/docs/integrations/providers/writer.mdx
View File

@ -51,3 +51,5 @@ Support of basic function calls defined via dicts, Pydantic, python functions et
```python ```python
from langchain_writer.tools import GraphTool from langchain_writer.tools import GraphTool
``` ```
Writer-specific remotely invoking tool

50
docs/docs/integrations/splitters/writer_text_splitter.ipynb
View File

@ -15,8 +15,8 @@
"\n", "\n",
"### Integration details\n", "### Integration details\n",
"| Class | Package | Local | Serializable | JS support | Package downloads | Package latest |\n", "| Class | Package | Local | Serializable | JS support | Package downloads | Package latest |\n",
"|:-----------------------------------------------------------------------------------------------------------------------------------------------------------|:-----------------| :---: | :---: |:----------:|:------------------------------------------------------------------------------------------------:|:---------------------------------------------------------------------------------------------:|\n", "|:-----------------------------------------------------------------------------------------------------------------------------------------|:-----------------| :---: | :---: |:----------:|:------------------------------------------------------------------------------------------------:|:---------------------------------------------------------------------------------------------:|\n",
"| WriterTextSplitter | [langchain-writer](https://pypi.org/project/langchain-writer/) | ❌ | ❌ | ❌ | ![PyPI - Downloads](https://img.shields.io/pypi/dm/langchain-writer?style=flat-square&label=%20) | ![PyPI - Version](https://img.shields.io/pypi/v/langchain-writer?style=flat-square&label=%20) |" "| [WriterTextSplitter](https://github.com/writer/langchain-writer/blob/main/langchain_writer/text_splitter.py#L11) | [langchain-writer](https://pypi.org/project/langchain-writer/) | ❌ | ❌ | ❌ | ![PyPI - Downloads](https://img.shields.io/pypi/dm/langchain-writer?style=flat-square&label=%20) | ![PyPI - Version](https://img.shields.io/pypi/v/langchain-writer?style=flat-square&label=%20) |"
] ]
}, },
{ {
@ -31,11 +31,11 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "a8d653f15b7ee32d", "id": "a8d653f15b7ee32d",
"metadata": {}, "metadata": {},
"source": "%pip install --quiet -U langchain-writer",
"outputs": [], "outputs": [],
"source": "%pip install --quiet -U langchain-writer" "execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -49,17 +49,17 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "2983e19c9d555e58", "id": "2983e19c9d555e58",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"import getpass\n", "import getpass\n",
"import os\n", "import os\n",
"\n", "\n",
"if not os.getenv(\"WRITER_API_KEY\"):\n", "if not os.getenv(\"WRITER_API_KEY\"):\n",
" os.environ[\"WRITER_API_KEY\"] = getpass.getpass(\"Enter your Writer API key: \")" " os.environ[\"WRITER_API_KEY\"] = getpass.getpass(\"Enter your Writer API key: \")"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -71,14 +71,14 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "98d8422ecee77403", "id": "98d8422ecee77403",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"# os.environ[\"LANGCHAIN_TRACING_V2\"] = \"true\"\n", "# os.environ[\"LANGCHAIN_TRACING_V2\"] = \"true\"\n",
"# os.environ[\"LANGCHAIN_API_KEY\"] = getpass.getpass()" "# os.environ[\"LANGCHAIN_API_KEY\"] = getpass.getpass()"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -96,15 +96,15 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "787b3ba8af32533f", "id": "787b3ba8af32533f",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"from langchain_writer.text_splitter import WriterTextSplitter\n", "from langchain_writer.text_splitter import WriterTextSplitter\n",
"\n", "\n",
"splitter = WriterTextSplitter(strategy=\"fast_split\")" "splitter = WriterTextSplitter(strategy=\"fast_split\")"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -120,10 +120,8 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "d1a24b81a8a96f09", "id": "d1a24b81a8a96f09",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"text = \"\"\"Reeeeeeeeeeeeeeeeeeeeeaally long text you want to divide into smaller chunks. For example you can add a poem multiple times:\n", "text = \"\"\"Reeeeeeeeeeeeeeeeeeeeeaally long text you want to divide into smaller chunks. For example you can add a poem multiple times:\n",
"Two roads diverged in a yellow wood,\n", "Two roads diverged in a yellow wood,\n",
@ -201,7 +199,9 @@
"\n", "\n",
"chunks = splitter.split_text(text)\n", "chunks = splitter.split_text(text)\n",
"chunks" "chunks"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -213,13 +213,13 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "a470daa875d99006", "id": "a470daa875d99006",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"print(len(chunks))" "print(len(chunks))"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -232,14 +232,14 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "e2f7fd52b7188c6c", "id": "e2f7fd52b7188c6c",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"async_chunks = await splitter.asplit_text(text)\n", "async_chunks = await splitter.asplit_text(text)\n",
"async_chunks" "async_chunks"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -251,13 +251,13 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "a1439db14e687fa4", "id": "a1439db14e687fa4",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"print(len(async_chunks))" "print(len(async_chunks))"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",

72
docs/docs/integrations/tools/writer.ipynb
View File

@ -15,8 +15,8 @@
"### Integration details\n", "### Integration details\n",
"\n", "\n",
"| Class | Package | Local | Serializable | JS support | Package downloads | Package latest |\n", "| Class | Package | Local | Serializable | JS support | Package downloads | Package latest |\n",
"|:-------------------------------------------------------------------------------------------------------------------------------------------|:-----------------| :---: | :---: |:----------:|:------------------------------------------------------------------------------------------------:|:---------------------------------------------------------------------------------------------:|\n", "|:-----------------------------------------------------------------------------------------------------------|:-----------------| :---: | :---: |:----------:|:------------------------------------------------------------------------------------------------:|:---------------------------------------------------------------------------------------------:|\n",
"| GraphTool | [langchain-writer](https://pypi.org/project/langchain-writer/) | ❌ | ❌ | ❌ | ![PyPI - Downloads](https://img.shields.io/pypi/dm/langchain-writer?style=flat-square&label=%20) | ![PyPI - Version](https://img.shields.io/pypi/v/langchain-writer?style=flat-square&label=%20) |\n", "| [GraphTool](https://github.com/writer/langchain-writer/blob/main/langchain_writer/tools.py#L9) | [langchain-writer](https://pypi.org/project/langchain-writer/) | ❌ | ❌ | ❌ | ![PyPI - Downloads](https://img.shields.io/pypi/dm/langchain-writer?style=flat-square&label=%20) | ![PyPI - Version](https://img.shields.io/pypi/v/langchain-writer?style=flat-square&label=%20) |\n",
"\n", "\n",
"### Features\n", "### Features\n",
"\n", "\n",
@ -43,17 +43,17 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "80d4e1a791aaa8", "id": "80d4e1a791aaa8",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"import getpass\n", "import getpass\n",
"import os\n", "import os\n",
"\n", "\n",
"if not os.getenv(\"WRITER_API_KEY\"):\n", "if not os.getenv(\"WRITER_API_KEY\"):\n",
" os.environ[\"WRITER_API_KEY\"] = getpass.getpass(\"Enter your Writer API key: \")" " os.environ[\"WRITER_API_KEY\"] = getpass.getpass(\"Enter your Writer API key: \")"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -77,10 +77,8 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "6faaae25509f0f28", "id": "6faaae25509f0f28",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"from langchain_writer.chat_models import ChatWriter\n", "from langchain_writer.chat_models import ChatWriter\n",
"from langchain_writer.tools import GraphTool\n", "from langchain_writer.tools import GraphTool\n",
@ -89,7 +87,9 @@
"\n", "\n",
"graph_id = getpass.getpass(\"Enter Writer Knowledge Graph ID: \")\n", "graph_id = getpass.getpass(\"Enter Writer Knowledge Graph ID: \")\n",
"graph_tool = GraphTool(graph_ids=[graph_id])" "graph_tool = GraphTool(graph_ids=[graph_id])"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -99,10 +99,8 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "e98d7deedb0e5c6f", "id": "e98d7deedb0e5c6f",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"from typing import Optional\n", "from typing import Optional\n",
"\n", "\n",
@ -154,7 +152,9 @@
" },\n", " },\n",
" },\n", " },\n",
"}" "}"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -167,15 +167,15 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "a4833f2597a87777", "id": "a4833f2597a87777",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"chat.bind_tools(\n", "chat.bind_tools(\n",
" [graph_tool, get_supercopa_trophies_count, GetWeather, get_product_info]\n", " [graph_tool, get_supercopa_trophies_count, GetWeather, get_product_info]\n",
")" ")"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -187,13 +187,13 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "ccb61b945a56672b", "id": "ccb61b945a56672b",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"chat.tools" "chat.tools"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -205,13 +205,13 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "381f0d4b9a8357a4", "id": "381f0d4b9a8357a4",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"chat.tool_choice" "chat.tool_choice"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -225,10 +225,8 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "74df06b58b5dc2e9", "id": "74df06b58b5dc2e9",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"from langchain_core.messages import HumanMessage\n", "from langchain_core.messages import HumanMessage\n",
"\n", "\n",
@ -240,7 +238,9 @@
"\n", "\n",
"response = chat.invoke(messages)\n", "response = chat.invoke(messages)\n",
"messages.append(response)" "messages.append(response)"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -252,13 +252,13 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "e271e0fc677446b2", "id": "e271e0fc677446b2",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"print(response.tool_calls)" "print(response.tool_calls)"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -270,10 +270,8 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "156b58108aa9b367", "id": "156b58108aa9b367",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"for tool_call in response.tool_calls:\n", "for tool_call in response.tool_calls:\n",
" selected_tool = {\n", " selected_tool = {\n",
@ -284,7 +282,9 @@
"\n", "\n",
"response = chat.invoke(messages)\n", "response = chat.invoke(messages)\n",
"print(response.content)" "print(response.content)"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -296,13 +296,13 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "4b3c6f05096fc9e3", "id": "4b3c6f05096fc9e3",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"print(response.additional_kwargs[\"graph_data\"])" "print(response.additional_kwargs[\"graph_data\"])"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -314,13 +314,13 @@
}, },
{ {
"cell_type": "code", "cell_type": "code",
"execution_count": null,
"id": "eb6e0da74b10b8fc", "id": "eb6e0da74b10b8fc",
"metadata": {}, "metadata": {},
"outputs": [],
"source": [ "source": [
"print(response.content)" "print(response.content)"
] ],
"outputs": [],
"execution_count": null
}, },
{ {
"cell_type": "markdown", "cell_type": "markdown",
@ -329,7 +329,7 @@
"source": [ "source": [
"## Chaining\n", "## Chaining\n",
"\n", "\n",
"#TODO: fill chaining section" "Due to specificity of Writer Graph tool (you don't need to call it manually, Writer server will call it by himself and return RAG based generation) it's impossible to invoke it separately, so GraphTool can't be used as part of chain"
] ]
}, },
{ {

2
libs/community/langchain_community/callbacks/bedrock_anthropic_callback.py
View File

@ -11,6 +11,7 @@ MODEL_COST_PER_1K_INPUT_TOKENS = {
"anthropic.claude-3-sonnet-20240229-v1:0": 0.003, "anthropic.claude-3-sonnet-20240229-v1:0": 0.003,
"anthropic.claude-3-5-sonnet-20240620-v1:0": 0.003, "anthropic.claude-3-5-sonnet-20240620-v1:0": 0.003,
"anthropic.claude-3-5-sonnet-20241022-v2:0": 0.003, "anthropic.claude-3-5-sonnet-20241022-v2:0": 0.003,
"anthropic.claude-3-7-sonnet-20250219-v1:0": 0.003,
"anthropic.claude-3-haiku-20240307-v1:0": 0.00025, "anthropic.claude-3-haiku-20240307-v1:0": 0.00025,
"anthropic.claude-3-opus-20240229-v1:0": 0.015, "anthropic.claude-3-opus-20240229-v1:0": 0.015,
"anthropic.claude-3-5-haiku-20241022-v1:0": 0.0008, "anthropic.claude-3-5-haiku-20241022-v1:0": 0.0008,
@ -23,6 +24,7 @@ MODEL_COST_PER_1K_OUTPUT_TOKENS = {
"anthropic.claude-3-sonnet-20240229-v1:0": 0.015, "anthropic.claude-3-sonnet-20240229-v1:0": 0.015,
"anthropic.claude-3-5-sonnet-20240620-v1:0": 0.015, "anthropic.claude-3-5-sonnet-20240620-v1:0": 0.015,
"anthropic.claude-3-5-sonnet-20241022-v2:0": 0.015, "anthropic.claude-3-5-sonnet-20241022-v2:0": 0.015,
"anthropic.claude-3-7-sonnet-20250219-v1:0": 0.015,
"anthropic.claude-3-haiku-20240307-v1:0": 0.00125, "anthropic.claude-3-haiku-20240307-v1:0": 0.00125,
"anthropic.claude-3-opus-20240229-v1:0": 0.075, "anthropic.claude-3-opus-20240229-v1:0": 0.075,
"anthropic.claude-3-5-haiku-20241022-v1:0": 0.004, "anthropic.claude-3-5-haiku-20241022-v1:0": 0.004,

2
libs/community/langchain_community/retrievers/tavily_search_api.py
View File

@ -123,7 +123,7 @@ class TavilySearchAPIRetriever(BaseRetriever):
Document( Document(
page_content=result.get("content", "") page_content=result.get("content", "")
if not self.include_raw_content if not self.include_raw_content
else result.get("raw_content", ""), else (result.get("raw_content") or ""),
metadata={ metadata={
"title": result.get("title", ""), "title": result.get("title", ""),
"source": result.get("url", ""), "source": result.get("url", ""),

9
libs/community/langchain_community/tools/tavily_search/tool.py
View File

@ -51,9 +51,12 @@ class TavilySearchResults(BaseTool): # type: ignore[override, override]
tool.invoke({'query': 'who won the last french open'}) tool.invoke({'query': 'who won the last french open'})
.. code-block:: python .. code-block:: json
'{\n "url": "https://www.nytimes.com...", "content": "Novak Djokovic won the last French Open by beating Casper Ruud ...' {
"url": "https://www.nytimes.com...",
"content": "Novak Djokovic won the last French Open by beating Casper Ruud ..."
}
Invoke with tool call: Invoke with tool call:
@ -64,7 +67,7 @@ class TavilySearchResults(BaseTool): # type: ignore[override, override]
.. code-block:: python .. code-block:: python
ToolMessage( ToolMessage(
content='{\n "url": "https://www.nytimes.com...", "content": "Novak Djokovic won the last French Open by beating Casper Ruud ...', content='{ "url": "https://www.nytimes.com...", "content": "Novak Djokovic won the last French Open by beating Casper Ruud ..." }',
artifact={ artifact={
'query': 'who won the last french open', 'query': 'who won the last french open',
'follow_up_questions': None, 'follow_up_questions': None,

87
libs/partners/anthropic/langchain_anthropic/chat_models.py
View File

@ -509,6 +509,93 @@ class ChatAnthropic(BaseChatModel):
"The image depicts a sunny day with a partly cloudy sky. The sky is a brilliant blue color with scattered white clouds drifting across. The lighting and cloud patterns suggest pleasant, mild weather conditions. The scene shows a grassy field or meadow with a wooden boardwalk trail leading through it, indicating an outdoor setting on a nice day well-suited for enjoying nature." "The image depicts a sunny day with a partly cloudy sky. The sky is a brilliant blue color with scattered white clouds drifting across. The lighting and cloud patterns suggest pleasant, mild weather conditions. The scene shows a grassy field or meadow with a wooden boardwalk trail leading through it, indicating an outdoor setting on a nice day well-suited for enjoying nature."
Extended thinking:
Claude 3.7 Sonnet supports an
`extended thinking <https://docs.anthropic.com/en/docs/build-with-claude/extended-thinking>`_
feature, which will output the step-by-step reasoning process that led to its
final answer.
To use it, specify the `thinking` parameter when initializing `ChatAnthropic`.
It can also be passed in as a kwarg during invocation.
You will need to specify a token budget to use this feature. See usage example:
.. code-block:: python
from langchain_anthropic import ChatAnthropic
llm = ChatAnthropic(
model="claude-3-7-sonnet-latest",
max_tokens=5000,
thinking={"type": "enabled", "budget_tokens": 2000},
)
response = llm.invoke("What is the cube root of 50.653?")
response.content
.. code-block:: python
[{'signature': '...', 'thinking': "To find the cube root of 50.653...", 'type': 'thinking'}, {'text': 'The cube root of 50.653 is ...', 'type': 'text'}]
Citations:
Anthropic supports a
`citations <https://docs.anthropic.com/en/docs/build-with-claude/citations>`_
feature that lets Claude attach context to its answers based on source
documents supplied by the user. When
`document content blocks <https://docs.anthropic.com/en/docs/build-with-claude/citations#document-types>`_
with ``"citations": {"enabled": True}`` are included in a query, Claude may
generate citations in its response.
.. code-block:: python
from langchain_anthropic import ChatAnthropic
llm = ChatAnthropic(model="claude-3-5-haiku-latest")
messages = [
{
"role": "user",
"content": [
{
"type": "document",
"source": {
"type": "text",
"media_type": "text/plain",
"data": "The grass is green. The sky is blue.",
},
"title": "My Document",
"context": "This is a trustworthy document.",
"citations": {"enabled": True},
},
{"type": "text", "text": "What color is the grass and sky?"},
],
}
]
response = llm.invoke(messages)
response.content
.. code-block:: python
[{'text': 'Based on the document, ', 'type': 'text'},
{'text': 'the grass is green',
'type': 'text',
'citations': [{'type': 'char_location',
'cited_text': 'The grass is green. ',
'document_index': 0,
'document_title': 'My Document',
'start_char_index': 0,
'end_char_index': 20}]},
{'text': ', and ', 'type': 'text'},
{'text': 'the sky is blue',
'type': 'text',
'citations': [{'type': 'char_location',
'cited_text': 'The sky is blue.',
'document_index': 0,
'document_title': 'My Document',
'start_char_index': 20,
'end_char_index': 36}]},
{'text': '.', 'type': 'text'}]
Token usage: Token usage:
.. code-block:: python .. code-block:: python

18
libs/partners/openai/tests/unit_tests/chat_models/test_azure.py
View File

@ -4,6 +4,7 @@ import os
from unittest import mock from unittest import mock
import pytest import pytest
from langchain_core.messages import HumanMessage
from typing_extensions import TypedDict from typing_extensions import TypedDict
from langchain_openai import AzureChatOpenAI from langchain_openai import AzureChatOpenAI
@ -81,3 +82,20 @@ def test_structured_output_old_model() -> None:
# assert tool calling was used instead of json_schema # assert tool calling was used instead of json_schema
assert "tools" in llm.steps[0].kwargs # type: ignore assert "tools" in llm.steps[0].kwargs # type: ignore
assert "response_format" not in llm.steps[0].kwargs # type: ignore assert "response_format" not in llm.steps[0].kwargs # type: ignore
def test_max_completion_tokens_in_payload() -> None:
llm = AzureChatOpenAI(
azure_deployment="o1-mini",
api_version="2024-12-01-preview",
azure_endpoint="my-base-url",
model_kwargs={"max_completion_tokens": 300},
)
messages = [HumanMessage("Hello")]
payload = llm._get_request_payload(messages)
assert payload == {
"messages": [{"content": "Hello", "role": "user"}],
"model": None,
"stream": False,
"max_completion_tokens": 300,
}