Media coverage
A Second Year of Building AI for Business Leaders
Crossing revisits Jomy to discuss Zleap’s evolution from a CEO management appliance to a Token Router and Zleap OS, unified enterprise AI, and continuously running agents.
Product
SAG Wiki: A Lightweight, Low-Cost, Precise Local Knowledge Base
Introducing SAG Wiki, the open-source knowledge base powered by SAG2, with folder-based organization, fast indexing and retrieval, and desktop and CLI access for people and agents.
Technology
Introducing SAG2: A New State of the Art, 7,000× Faster Indexing, No Vectors, Just Files
SAG2 combines keyword indexing with dynamic multi-hop retrieval. Explore indexing performance, public benchmarks, SAG Wiki and applications in agent memory.
Media coverage
A Second Year of Building AI for Business Leaders
Crossing revisits Jomy to discuss Zleap’s evolution from a CEO management appliance to a Token Router and Zleap OS, unified enterprise AI, and continuously running agents.
Media coverage
A New RAG State of the Art, with Second-Scale Retrieval Across 500 Million Records
Synced examines SAG’s event-and-entity structure, SQL and semantic retrieval, benchmark results, and production deployments for agent knowledge retrieval.
Media coverage
The Next Phase of Agents: Building the Infrastructure for Enterprise AI in China
Z Potentials examines local deployment, costs, enterprise data, SAG, Zleap-Agent, and collaboration between enterprise agents through Zleap’s approach.
Media coverage
CEOs, Your Next Personal Assistant Need Not Be Human
QbitAI tries Zleap’s agent appliance, exploring information collection, automated reports, source tracing, report-specific Q&A, and Jomy’s product strategy.
Media coverage
An Executive Briefcase for Every CEO, for RMB 68,000
Xinzhiyuan reports on Zleap-D1, exploring CEO-focused information analysis, mobile use, local deployment, and multi-agent collaboration.
Media coverage
An Agent Appliance Built for CEOs: A Conversation with 302.AI Founder Jomy
Crossing interviews Jomy about enterprise AI through CEO information needs, local deployment, small-model engineering, and traceable knowledge retrieval.