Nace AI Launches New Drex for Search Tool
The system acts as a retrieval layer for various file formats to improve search accuracy for AI agents.
Updated on Oct. 26, 2026 in Artificial Intelligence
Live Poll
Do you trust current AI tools to accurately find and cite information from large document collections?
Nace AI has released Drex for Search, a tool designed to serve as a retrieval layer for documents and agents. The system addresses limitations in classic retrieval-augmented generation pipelines, such as handling table data and multi-hop requirements.
Why it matters
This release targets shortcomings in existing RAG pipelines including scanned PDF processing and complex table data analysis. It enables better integration between AI agents and diverse enterprise file types.
The system achieves a 94.7% accuracy rate on the MADQA benchmark using Gemini 3.5 Flash and scores 88.22 on the EnterpriseRAG-Bench. It supports multiple file types including PDF, xlsx, csv, Parquet, mp3, mp4, docx, pptx, md, json, and txt.
The players
Nace AI
Nace AI is a technology firm that specializes in developing retrieval systems and AI agents for enterprise data environments.
The details
Drex for Search classifies queries to determine if a quick lookup or an agent search is necessary. It narrows focus across multiple levels, including folders, files, sections, pages, tables, and columns to optimize retrieval.
Timeline
October 20, 2026: Nace AI reviewed public leaderboards for benchmark rankings.
October 26, 2026: Nace AI released Drex for Search.
The Tech Race
This release marks a significant effort to replace the limitations found in existing legacy RAG pipelines with a more granular retrieval layer. By outperforming established tools, Nace AI positions its software as a more capable alternative for handling complex enterprise data workflows.
Developers and coding agents can access the system via CLI, SDKs, MCP, or REST to improve data retrieval capabilities. The tool is compatible with managed cloud, AWS, Azure, and on-premise environments, offering flexibility for different technical setups.
The takeaway
Nace AI is attempting to bridge the gap in document retrieval by categorizing search tasks more effectively than previous methods. Users should focus on how the multi-level search function could streamline their specific file and table data retrieval workflows.
Further reading
For more on industry developments, see our coverage of Artificial Intelligence.
Live Poll
Do you trust current AI tools to accurately find and cite information from large document collections?
