ReadingResearchRadarInvestment framework
Sign in / Sign up
← Back to reading
TopicDisconfirmAI EditorialUpdated 2026-11-22 07:42

Norizon AI Funding

2 steps1 pieces4 Sources

Beijing-based AI startup Norizon AI (Wuya Zhiyuan) has completed angel and angel-plus funding rounds totaling several hundred million RMB. The company focuses on Real-Time Recursive Self-Improvement (RSI), a technology designed to allow large language models to continuously learn and update their own parameters after deployment, rather than relying solely on context windows or external retrieval.

The funding round closed on November 19, 2026, with participation from investors including 5Y Capital, Sequoia China (HongShan), Lenovo Star, Zhongguancun Science City Fund, and Gobi Partners. FlowCapital served as the financial advisor. The proceeds will be used for R&D and product commercialization of their proprietary SHINE hypernetwork technology (zhidx.com ↗).

Norizon AI's core technical approach involves using a pretrained hypernetwork to generate lightweight parameter updates in a single forward pass. This aims to reduce the computational cost of continuous learning to near-inference levels while mitigating catastrophic forgetting. The company claims its method outperforms fine-tuning controls in retaining general-task performance while learning new knowledge, though independent benchmark figures have not been publicly released (ai-market-watch.com ↗).

The significance of this development lies in the shift from static model versions to dynamic, self-evolving systems. While the economic feasibility of broad recursive self-improvement remains debated in academic circles (arxiv.org ↗), Norizon AI is positioning itself as a practical implementer of narrow, task-specific parameter updates. Key challenges remain in verifying long-term stability across millions of tokens and proving that these updates do not degrade base capabilities over time.

Readers should watch for: 1) Public release of technical benchmarks comparing SHINE against standard fine-tuning; 2) Announcements of pilot deployments in enterprise or agent environments; 3) Further details on the team's research into scaling hypernetwork size and internalization length.

Sign in / Sign up
ReadingResearchRadarInvestment framework

How it got here

  1. 2026-11-19

    November 19, 2026: Norizon AI announces completion of angel and angel-plus rounds totaling several hundred million RMB. Investors include 5Y Capital, Sequoia China, Lenovo Star, Zhongguancun Science City Fund, and Gobi Partners. ([zhidx.com](https://zhidx.com/p/612275.html))

    zhidx.com ↗
  2. 2026-11-19

    November 19, 2026: Technical details emerge regarding the SHINE hypernetwork, which generates parameter updates in a single forward pass to enable real-time self-improvement without repeated gradient optimization. ([elsewhere.news](https://elsewhere.news/en/5ycapital/rsi-5y-news))

    elsewhere.news ↗

What is still unsettled

  • What are the specific quantitative benchmarks demonstrating SHINE's superiority over fine-tuning in preventing catastrophic forgetting?
  • How does Norizon AI plan to address the stability of parameter updates over long sequences (millions of tokens) in production environments?
  • Which specific enterprise or consumer products will first integrate this RSI technology?

Sources

  • zhidx.comZhidongxi Report↗
  • elsewhere.news5Y Capital News↗
  • ai-market-watch.comAI Market Watch Analysis↗
  • arxiv.orgArXiv Paper on RSI Economics↗

Coverage in this topic

  • Beijing AI Startup Norizon AI Completes Hundreds of Millions RMB Angel and Angel+ Rounds2026-11-19News

    Norizon AI, focused on Real-Time Recursive Self-Improvement (RSI), secures funding from Five Yields Capital, Sequoia China, and others to advance model continuous learning technology.