{"repo":"lotus-data/lotus","free":true,"listed":false,"github":"https://github.com/lotus-data/lotus","clone":"git clone https://github.com/lotus-data/lotus.git","description":"Optimized Agentic and LLM Bulk Processing Over Your Data","language":"Python","stars":1658,"topics":["ai-data-processing","data","llm","pandas","python","semantic-operators","semantic-search","unstructured-data","llm-data-processing","llm-document-processing"],"license":"Apache-2.0","category":"ai-agents","readme_excerpt":"LOTUS: Optimized Agentic and LLM Bulk Processing Bulk process your datasets with agents and LLMs at scale, with higher accuracy and lower cost. From Stanford University and UC Berkeley [ ][#pypi-package] [ ][#pypi-package] [ ][#arxiv-paper-package] [ ][#discord] What is LOTUS? • Install • Quickstart • Semantic Operators • Community • Docs • Cite --- What is LOTUS? LOTUS makes agentic and LLM bulk processing fast, easy, and robust. It introduces and optimizes semantic operators (e.g., LLM-based map , reduce , filter primitives) to let you process your large datasets with LLMs and natural language instructions. LOTUS optimizes these operations to help you get higher accuracy and lower cost . What you can build: - Agentic code processing — run a tool-using agent (with a sandboxed Python REPL) over every file, document, or record, then reduce to one answer (codebase analysis, security sweeps, migrations). [example] - Deep research & synthesis — fan out over a corpus, extract, and synthesize. [example, blog] - Agent-trace failure analysis — mine large volumes of agent logs for failure modes. [blog] - Document extraction & unstructured analysis — structured fields and insights from text. [example, blog] - LLM-judge evals — score model outputs, app responses, and content quality, row by row. [example, blog] - RAG — retrieve the relevant context and generate grounded answers over your data. [blog] LOTUS stands for L LMs O ver T ext, U nstructured and S tructured Data. Installation Or","default_branch":null,"files":null,"tree":[],"storefront":"/r/lotus-data","claimed":false,"request_supported":{"post":"https://gitbuyer.com/r/lotus-data/lotus/request-supported","requests":0},"note":"indexed from public GitHub; nothing is for sale on this page. Clone it from GitHub. Paid listings live at /search."}