langchain-ai · ernie-c-jeong · Sep 17, 2025 · Sep 17, 2025
diff --git a/docs/docs/integrations/document_loaders/example_data/polaris_ai_example.docx b/docs/docs/integrations/document_loaders/example_data/polaris_ai_example.docx
diff --git a/docs/docs/integrations/document_loaders/polaris_ai_datainsight.ipynb b/docs/docs/integrations/document_loaders/polaris_ai_datainsight.ipynb
@@ -0,0 +1,245 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "433f5422ad8e1efa",
+   "metadata": {
+    "collapsed": false,
+    "jupyter": {
+     "outputs_hidden": false
+    }
+   },
+   "source": [
+    "# Polaris AI DataInsight\n",
+    "\n",
+    "> [Polaris AI DataInsight](https://datainsight.polarisoffice.com/playground) is a document parser that extracts document elements (text, images, complex tables, charts, etc.) from various file formats into structured JSON, making them easy to integrate into RAG systems.\n",
+    "\n",
+    "This notebook covers how to get started with `PolarisAIDatainsightLoader`.\n",
+    "\n",
+    "\n",
+    "## Installation\n",
+    "\n",
+    "Install `langchain-polaris-ai-datainsight` package."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "0ae97af4",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\n",
+      "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m A new release of pip is available: \u001b[0m\u001b[31;49m25.1.1\u001b[0m\u001b[39;49m -> \u001b[0m\u001b[32;49m25.2\u001b[0m\n",
+      "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m To update, run: \u001b[0m\u001b[32;49mpip install --upgrade pip\u001b[0m\n",
+      "Note: you may need to restart the kernel to use updated packages.\n"
+     ]
+    }
+   ],
+   "source": [
+    "%pip install -qU --pre langchain-polaris-ai-datainsight"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "e6e5941c",
+   "metadata": {},
+   "source": [
+    "## Environment Setup\n",
+    "\n",
+    "Make sure to set the following environment variables:\n",
+    "\n",
+    "- `POLARIS_AI_DATA_INSIGHT_API_KEY`: Your Polaris AI DataInsight API key. Read [Polaris AI DataInsight Documentation](https://datainsight.polarisoffice.com/api/keys) to get your API key."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "21e72f3d",
+   "metadata": {},
+   "source": [
+    "## Usage"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "a05efd34",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import getpass\n",
+    "import os\n",
+    "\n",
+    "os.environ[\"POLARIS_AI_DATA_INSIGHT_API_KEY\"] = getpass.getpass(\n",
+    "    \"Enter your PolarisAIDataInsight API key: \"\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "2b914a7b",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      " --------- < Page Content > --------- \n",
+      "2025 Seed Program Application \n",
+      "\n",
+      "I. Funding Information by Track\n",
+      "\n",
+      "1. Beginning and Advanced Track Comparison Overview\n",
+      "\n",
+      "<table><tbody><tr><td>Category</td><td>Beginning Track*</td><td>Advanced Track*</td></tr><tr><td>Funding target</td><td>A university located outside Korea that has a Central Grant Management Department, an existing Korean Studies infrastructure, and plans to establish an education foundation.</td><td>A non-Korean university with a Central Grant Management Department, at least one full-time Korean Studies faculty member, an undergraduate Korean Studies major or department, and commitment to supporting Korean Studies.</td></tr><tr><td>Funding period</td><td>3 years</td><td>5 years<3+2years></td></tr><tr><td>Funding size</td><td>Maximum possible funding depends on the applicant university’s country<br><table><tbody><tr><td>Country Group*</td><td>Maximum Funding**</td></tr><tr><td>A</td><td>Up to KRW 200 million</td></tr><tr><td>B</td><td>Up to KRW 50 million</td></tr></tbody></table></td><td>Maximum possible funding depends on the applicant university’s country<br><table><tbody><tr><td>Country Group*</td><td>Maximum Funding**</td></tr><tr><td>A</td><td>Up to KRW 150 million</td></tr><tr><td>B</td><td>Up to KRW 90 million</td></tr></tbody></table></td></tr><tr><td>Required project content</td><td>·\tFund 2 or more scholarship students<br>·\tOffer 1 or more regular Korean Studies lecture courses (Excluding Korean language courses)<br>·\tHold 1 or more workshops per year in which that students may participate</td><td>·\tHire 1 or more Korean Studies full-time faculty<br>·\tFund 1 or more scholarship student for Korean Studies<br>·\tOffer 2 or more regular graduate-level Korean Studies lecture courses (Excluding Korean language courses)<br>·\tHold 1 or more international Korean Studies conference<br>·\tEstablish and manage a website, blog, or social media relating to the program </td></tr><tr><td>Recommended content</td><td>·\tFoster talent (education)<br>·\tEstablish a Korean Studies research institute/center<br>·\tEstablish Korean Studies undergraduate department/major & program<br>·\tDevelop Korean Studies textbooks<br>·\tHold academic activities</td><td>·\tFoster talent (education)<br>·\tEstablish a Korean Studies research institute/center<br>·\tEstablish Korean Studies M.A/Ph.D. department/major & program<br>·\tDevelop Korean Studies textbooks<br>·\tHold academic activities</td></tr></tbody></table>\n",
+      "\n",
+      "<img id=\"di.image.im12\" data-category=\"image\"/>\n",
+      "\n",
+      " 2 / 3\n",
+      "\n",
+      "\n",
+      " --------- < Metadata > --------- \n",
+      "{'di.text.he2te0': {'id': 'di.text.he2te0', 'type': 'text'}, 'di.text.te0': {'id': 'di.text.te0', 'type': 'text'}, 'di.text.te2': {'id': 'di.text.te2', 'type': 'text'}, 'di.table.ta9': {'id': 'di.table.ta9', 'type': 'table'}, 'di.image.im12': {'id': 'di.image.im12', 'type': 'image', 'src': '/home/jenkins_agent/Project/langchain/docs/docs/integrations/document_loaders/example_data/tmp/tmpaynkptxx/polaris_ai_example.docx_image12.png'}, 'di.text.fo3te0': {'id': 'di.text.fo3te0', 'type': 'text'}}\n",
+      "\n",
+      "\n",
+      " --------- < Page Content > --------- \n",
+      "2025 Seed Program Application \n",
+      "\n",
+      "II. Review and Selection \n",
+      "\n",
+      "1. Review Process\n",
+      "\n",
+      "<img id=\"di.image.im13\" data-category=\"image\"/>\n",
+      "\n",
+      "\n",
+      "\n",
+      "\n",
+      "\n",
+      "Review of whether the basic requirements for application have been met\n",
+      "\n",
+      "\n",
+      "\n",
+      "\n",
+      "\n",
+      "\n",
+      "\n",
+      "Review of the Project Proposal\n",
+      "\n",
+      "Admistered by the Expert Review Team\n",
+      "\n",
+      "\n",
+      "\n",
+      "\n",
+      "\n",
+      "\n",
+      "\n",
+      "Final review and decision\n",
+      "\n",
+      "Admistered by the Comprehensive Review Committee\n",
+      "\n",
+      "\n",
+      "\n",
+      "1. Preliminary Review\n",
+      "\n",
+      "\n",
+      "\n",
+      "2. Content Review (80 pts)\n",
+      "\n",
+      "\n",
+      "\n",
+      "3. Comprehensive Review (20 pts)\n",
+      "\n",
+      "2. Review Stages and Content\n",
+      "\n",
+      "Stage 1: Preliminary Review\n",
+      "\n",
+      "Conducted by Main Department\n",
+      "\n",
+      "●\tVerifies document submission, eligibility, and overlapping support.\n",
+      "\n",
+      "●\tApplications missing required documents, signatures, or failing to meet eligibility do not proceed.\n",
+      "\n",
+      "●\tApplications with Indirect Expenses over 10% of Direct Expenses (including Labor Expenses) are rejected.\n",
+      "\n",
+      "Stage 2: Content Review\n",
+      "\n",
+      "Conducted by Expert Review Team\n",
+      "\n",
+      "●\tOnline review: Points given individually\n",
+      "\n",
+      "●\tPanel review: Points determined by consensus\n",
+      "\n",
+      "●\tAssesses leadership potential, capacity, and project plans.\n",
+      "\n",
+      "●\tItems and scores assigned for evaluation.\n",
+      "\n",
+      "<table><tbody><tr><td>Areas</td><td>Items (Points)</td><td>Content</td></tr></tbody></table>\n",
+      "\n",
+      " 2 / 3\n",
+      "\n",
+      "\n",
+      " --------- < Metadata > --------- \n",
+      "{'di.text.he2te0': {'id': 'di.text.he2te0', 'type': 'text'}, 'di.text.te10': {'id': 'di.text.te10', 'type': 'text'}, 'di.text.te12': {'id': 'di.text.te12', 'type': 'text'}, 'di.image.im13': {'id': 'di.image.im13', 'type': 'image', 'src': '/home/jenkins_agent/Project/langchain/docs/docs/integrations/document_loaders/example_data/tmp/tmpaynkptxx/polaris_ai_example.docx_image13.png'}, 'di.text.sh15': {'id': 'di.text.sh15', 'type': 'text'}, 'di.text.sh16': {'id': 'di.text.sh16', 'type': 'text'}, 'di.text.sh16te0': {'id': 'di.text.sh16te0', 'type': 'text'}, 'di.text.sh17': {'id': 'di.text.sh17', 'type': 'text'}, 'di.text.sh18': {'id': 'di.text.sh18', 'type': 'text'}, 'di.text.sh19': {'id': 'di.text.sh19', 'type': 'text'}, 'di.text.sh19te0': {'id': 'di.text.sh19te0', 'type': 'text'}, 'di.text.sh19te1': {'id': 'di.text.sh19te1', 'type': 'text'}, 'di.text.sh20': {'id': 'di.text.sh20', 'type': 'text'}, 'di.text.sh21': {'id': 'di.text.sh21', 'type': 'text'}, 'di.text.sh22': {'id': 'di.text.sh22', 'type': 'text'}, 'di.text.sh22te0': {'id': 'di.text.sh22te0', 'type': 'text'}, 'di.text.sh22te1': {'id': 'di.text.sh22te1', 'type': 'text'}, 'di.text.sh23': {'id': 'di.text.sh23', 'type': 'text'}, 'di.text.sh23te0': {'id': 'di.text.sh23te0', 'type': 'text'}, 'di.text.sh24': {'id': 'di.text.sh24', 'type': 'text'}, 'di.text.sh24te0': {'id': 'di.text.sh24te0', 'type': 'text'}, 'di.text.sh25': {'id': 'di.text.sh25', 'type': 'text'}, 'di.text.sh25te0': {'id': 'di.text.sh25te0', 'type': 'text'}, 'di.text.te15': {'id': 'di.text.te15', 'type': 'text'}, 'di.text.te16': {'id': 'di.text.te16', 'type': 'text'}, 'di.text.te17': {'id': 'di.text.te17', 'type': 'text'}, 'di.text.te18': {'id': 'di.text.te18', 'type': 'text'}, 'di.text.te19': {'id': 'di.text.te19', 'type': 'text'}, 'di.text.te20': {'id': 'di.text.te20', 'type': 'text'}, 'di.text.te21': {'id': 'di.text.te21', 'type': 'text'}, 'di.text.te22': {'id': 'di.text.te22', 'type': 'text'}, 'di.text.te23': {'id': 'di.text.te23', 'type': 'text'}, 'di.text.te24': {'id': 'di.text.te24', 'type': 'text'}, 'di.text.te25': {'id': 'di.text.te25', 'type': 'text'}, 'di.text.te26': {'id': 'di.text.te26', 'type': 'text'}, 'di.table.ta26': {'id': 'di.table.ta26', 'type': 'table'}, 'di.text.fo3te0': {'id': 'di.text.fo3te0', 'type': 'text'}}\n",
+      "\n",
+      "\n",
+      " --------- < Page Content > --------- \n",
+      "2025 Seed Program Application \n",
+      "\n",
+      "<table><tbody><tr><td rowspan=\"3\">Evaluation of the Basis for the Project (40)</td><td>Potential to lead Korean Studies (20)</td><td>- Assess whether the university has a distinguished reputation in terms of history and academic disciplines.<br>- Evaluate the strength of the network between the Project Director and local researchers.</td></tr><tr><td>Performance capacity (20)<br>Eligibility criteria (10)</td><td>- Determine if the project director possesses the skills and commitment to execute the project (e.g., Korean language proficiency, influence within the institution, management skills).<br>- Review the achievements of collaborative researchers in Korean Studies.<br>- Confirm whether personnel (Beginning/Advanced) or coursework (Advanced) meet eligibility criteria.</td></tr><tr><td>University support (10)</td><td>- Measure the institution's willingness to support Korean Studies (financial, spatial, and human resources, appropriate indirect expense ratio).<br>- Assess the competency of the Central Grant Management Department.</td></tr><tr><td rowspan=\"2\">Evaluation of the Project Content (40)</td><td>Project plans (30)</td><td>- Ensure that the project objectives are realistic and well-defined.<br>- Verify that the plan aligns with local conditions.<br>- Review the suitability of the Project Team’s structure.<br>- Assess whether the budget plan reflects local price levels.</td></tr></tbody></table>\n",
+      "\n",
+      " 2 / 3\n",
+      "\n",
+      "\n",
+      " --------- < Metadata > --------- \n",
+      "{'di.text.he2te0': {'id': 'di.text.he2te0', 'type': 'text'}, 'di.table.ta29': {'id': 'di.table.ta29', 'type': 'table'}, 'di.text.fo3te0': {'id': 'di.text.fo3te0', 'type': 'text'}}\n",
+      "\n",
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "from langchain_polaris_ai_datainsight import PolarisAIDataInsightLoader\n",
+    "\n",
+    "loader = PolarisAIDataInsightLoader(\n",
+    "    file_path=\"example_data/polaris_ai_example.docx\",\n",
+    "    resources_dir=\"example_data/tmp\",\n",
+    "    mode=\"page\",  # \"element\", \"page\", or \"single\". (default is \"single\")\n",
+    ")\n",
+    "\n",
+    "docs = loader.load()  # or loader.lazy_load()\n",
+    "\n",
+    "for doc in docs[:3]:\n",
+    "    print(\" --------- < Page Content > --------- \")\n",
+    "    print(doc.page_content)\n",
+    "    print(\" --------- < Metadata > --------- \")\n",
+    "    print(doc.metadata)\n",
+    "    print(\"\\n\")"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "langchain-monorepo-py3.11",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.11.11"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
diff --git a/docs/docs/integrations/providers/polaris_ai_datainsight.mdx b/docs/docs/integrations/providers/polaris_ai_datainsight.mdx
@@ -0,0 +1,19 @@
+# Polaris AI DataInsight
+
+> [Polaris AI DataInsight](https://datainsight.polarisoffice.com/playground) is a document parser 
+> that extracts document elements (text, images, complex tables, charts, etc.) from various file formats 
+> into structured JSON, making them easy to integrate into RAG systems.
+
+## Installation and Setup
+
+```bash
+pip install langchain-polaris-ai-datainsight
+```
+
+## Document Loader
+
+See a [usage example](/docs/docs/integrations/document_loaders/polaris_ai_datainsight).
+
+```
+from langchain_polaris_ai_datainsight import PolarisAIDataInsightLoader 
+```
diff --git a/docs/src/theme/FeatureTables.js b/docs/src/theme/FeatureTables.js
@@ -1052,6 +1052,12 @@ const FEATURE_TABLES = {
                 source: "Various file types (see https://ds4sd.github.io/docling/)",
                 apiLink: "https://python.langchain.com/docs/integrations/document_loaders/docling/"
             },
+            {
+                name: "PolarisAIDataInsightLoader",
+                link: "../../integrations/document_loaders/polaris_ai_datainsight",
+                source: "Various file types (see https://datainsight.polarisoffice.com/documentation?docType=doc_extract)",
+                apiLink: "https://python.langchain.com/docs/integrations/document_loaders/polaris_ai_datainsight/"
+            },
         ]
     },
     vectorstores: {

diff --git a/libs/packages.yml b/libs/packages.yml
@@ -751,4 +751,8 @@ packages:
   path: libs/zeusdb
 - name: langchain-scraperapi
   path: .
-  repo: scraperapi/langchain-scraperapi
+  repo: scraperapi/langchain-scraperapi
+- name: langchain-polaris-ai-datainsight
+  path: .
+  repo: PolarisOffice/langchain-polaris-ai-datainsight
+  name_title: Polaris AI Data Insight