Nace AI has launched Drex for Search, a retrieval layer designed to enhance document search capabilities for both humans and AI agents. This system integrates with various coding agents and LLMs, including GPT-5 and Gemini 3.5 Flash, and has demonstrated strong performance on benchmarks like BrowseComp-Plus, achieving 92.05% accuracy with significantly fewer search calls than competitors. Drex for Search aims to improve upon traditional RAG methods by intelligently navigating complex document structures and multi-hop queries. AI
IMPACT Enhances AI agent capabilities in document retrieval and complex query resolution, potentially improving efficiency and accuracy in enterprise search applications.
RANK_REASON This is a product release from a company that is not a tier-1 frontier model lab, focused on improving retrieval for AI agents.
- BrowseComp-Plus
- Claude Code
- Claude Opus-5
- .codex
- Cursor+
- Drex
- Drex for Search
- EnterpriseRAG-Bench
- Gemini 3.5 Flash
- GPT-5
- Nace.AI
- OfficeQA Pro
- OpenAI File Search
- openCode
- Qwen 27B
- VisDocAgentBench
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →