China's Palantir-inspired Yuedian raises $4.5M Pre-A for enterprise AI
What's the deal? Beijing Yuedian TechnologyDealroom has a profile for this one. Try Dealroom →, an enterprise AI company that has long modelled itself on Palantir, has raised a 30,000,000 CNY (about $4.5 million) Pre-A round. Yingfutaike led, with Shuimu Venture CapitalDealroom has a profile for this one. Try Dealroom → and innofundDealroom has a profile for this one. Try Dealroom → participating.
What's the endgame? Yuedian builds decision intelligence for private, on-premise enterprise environments, letting AI handle cross-system data queries, business judgement, and task execution. Founder and chief executive officer Ren Xinqi wants the software to replace the "all-knowing veteran" inside a company.
Why Palantir? The company grew out of Minglue Group's knowledge graph product line and keeps Palantir as a product reference. "We are not moving Palantir's business map to China," Ren told CV Info, saying it retains the approach of using private data to support complex analysis, decisions, and execution while solving Chinese enterprises' own problems.
By the numbers: The four-year-old company has reached break-even, with nearly 100 paying customers, most of them large enterprises. Order and revenue compound annual growth topped 300% from 2023 to 2025, and first-half 2026 revenue already exceeded all of 2025. Ren expects revenue to pass 100 million CNY in 2026.
How it works: Yuedian uses an ontology to organise business objects — customers, orders, products, equipment — then deploys agents to handle queries, task breakdown, and execution. In one quality-tracing workflow, a process that once took several people multiple days across systems now runs in five minutes with one employee. At a panel maker, Yuedian's Knora-AI product lifted yield by 3‰, worth more than 20 million CNY.
What's the money for? Yuedian will spend on training its AIGO model, advancing its AI FDE approach, refining a recursive self-improvement (RSI) loop, and replicating validated use cases. Ren's near-term goal is a closed loop in some production settings that runs without continuous human involvement.
The signal: While many enterprise AI firms still struggle with proof-of-concept conversion and renewals, Yuedian's profitability and growth suggest demand for AI that not only answers questions but executes across siloed enterprise systems.
Image credit: infomatique
