---
title: "See you at Spring Festival? DeepSeek's next-generation model: \"High cost-performance\" innovative architecture to help China break through the bottleneck of \"computing power chips and memory\""
type: "News"
locale: "en"
url: "https://longbridge.com/en/news/275541677.md"
description: "Nomura Securities believes that DeepSeek's upcoming new generation large model V4 may further reduce training and inference costs through the innovative architecture mHC and Engram technology, accelerating the innovation cycle of China's AI value chain. It is also expected to help global large language model and AI application companies accelerate their commercialization process, thereby alleviating the increasingly heavy capital expenditure pressure"
datetime: "2026-02-11T02:40:21.000Z"
locales:
  - [zh-CN](https://longbridge.com/zh-CN/news/275541677.md)
  - [en](https://longbridge.com/en/news/275541677.md)
  - [zh-HK](https://longbridge.com/zh-HK/news/275541677.md)
generator: "portal-rs"
---

# See you at Spring Festival? DeepSeek's next-generation model: "High cost-performance" innovative architecture to help China break through the bottleneck of "computing power chips and memory"

Author: Bao Yilong

Source: Hard AI

Nomura Securities pointed out that DeepSeek's upcoming new generation large model V4 is expected not to trigger a global panic in AI computing power demand like last year's V3. However, it may accelerate the commercialization process of global large language AI applications through two underlying architectural innovations.

Wallstreet News mentioned that the new generation flagship model V4 is expected to be launched in mid-February 2026. Preliminary internal tests indicate that V4 surpasses other top models currently on the market, such as Anthropic's Claude and OpenAI's GPT series, in programming capabilities.

A core question arises again: **Will V4 disrupt the global AI value chain again?** Nomura Securities provided a clear judgment in its "Global AI Trend Tracking" report released on February 10: No.

**The research report pointed out that the significance of this release lies in V4's potential to further reduce training and inference costs through innovative architectures (mHC and Engram technology), accelerating the innovation cycle of China's AI value chain.**

**It is also expected to help global large language model and AI application companies accelerate the commercialization process, thereby alleviating the increasingly heavy capital expenditure pressure.**

## Innovative Technical Architecture Brings Performance and Cost Optimization

The report pointed out that computing chips and memory have always been bottlenecks for China's large models. The two key technologies that V4 is expected to introduce—mHC and Engram—optimize these hard constraints from algorithmic and engineering perspectives.

> **mHC**:
> 
> -   Full name "Manifold Constraint Hyperconnection." It aims to address the bottleneck of information flow and training instability in Transformer models when the number of layers is extremely deep.
> -   In simple terms, it makes the "dialogue" between neural network layers richer and more flexible, while preventing information from being amplified or destroyed through strict mathematical "guardrails." **Experiments have shown that models using mHC perform better in tasks such as mathematical reasoning.**
> 
> **** (Hyperconnection vs. Manifold Constraint Hyperconnection)
> 
> **Engram**:
> 
> -   A "conditional memory" module. Its design concept is to decouple "memory" from "computation."
> -   Static knowledge in the model (such as entities and fixed expressions) is specifically stored in a sparse memory table, which can be placed in inexpensive DRAM. When inference is needed, it is quickly retrieved. **This frees up expensive GPU memory (HBM), allowing it to focus on dynamic computation.**
> 
> **** (Engram architecture) **The research report points out that the combination of these two technologies is of great significance for the development of AI in China**. A more stable training process (mHC) compensates for potential shortcomings of domestic chips; smarter memory scheduling (Engram) bypasses the limitations of HBM capacity and bandwidth.

**Nomura emphasizes that the most direct commercial impact of V4 is to further reduce the training and inference costs of large models, and this cost-effectiveness improvement will stimulate demand, benefiting Chinese AI hardware companies with an accelerated investment cycle.**

## Hardware Benefits from the "Acceleration Cycle"

Nomura believes that major global cloud service providers are fully pursuing general artificial intelligence, and the competition for capital expenditure is far from over. Therefore, V4 is not expected to cause the same level of shockwaves in the global AI infrastructure market as last year.

**However, global large model and application developers are burdened with an increasingly heavy capital expenditure load. If V4 can significantly reduce training and inference costs while maintaining high performance as expected, it will serve as a strong boost.**

**It may help these players convert technology into revenue more quickly, alleviating profit pressure.**

The report reviews the market landscape one year after the release of DeepSeek-V3/R1.

Previously, the "computing power management efficiency" combined with "performance improvement" of DeepSeek's two models V3 and R1 accelerated the development of Chinese LLMs and applications, changed the competitive landscape of global and Chinese large language models, and increased attention to open-source models.

 (The weekly Token consumption of the top 15 open-source models on OpenRouter)

By the end of 2024, DeepSeek's two models accounted for more than half of the Token usage of open-source models on OpenRouter. However, by the second half of 2025, as more players joined, their market share had significantly declined.

The market has shifted from "one dominant player" to "a multitude of contenders." This indicates that the efficiency of a single model is no longer sufficient to dominate the rapidly evolving open-source ecosystem, and the competitive environment that V4 faces today is much more complex than a year ago.

## Software May Welcome "Value Addition Rather Than Replacement"

On the application side, the more powerful and efficient V4 will give rise to more robust AI agents.

The report observes that applications like Alibaba's Tongyi Qianwen App are already able to execute multi-step tasks in a more automated manner. This means that AI agents are transforming from "dialogue tools" into "AI assistants" capable of handling complex tasks.

These multi-tasking agents will need to interact more frequently with the underlying large models, which will consume more Tokens and subsequently increase computing power demand.

**Therefore, the improvement in model efficiency will not "kill software," but rather create value for leading software companies.**

Nomura emphasizes the need to focus on software companies that can be the first to leverage the capabilities of the new generation of large models to create disruptive AI-native applications or agents. Their growth ceiling may be raised again due to the leap in model capabilities

### Related Stocks

- [159995.CN](https://longbridge.com/en/quote/159995.CN.md)
- [512480.CN](https://longbridge.com/en/quote/512480.CN.md)
- [588170.CN](https://longbridge.com/en/quote/588170.CN.md)
- [512760.CN](https://longbridge.com/en/quote/512760.CN.md)
- [588780.CN](https://longbridge.com/en/quote/588780.CN.md)
- [159325.CN](https://longbridge.com/en/quote/159325.CN.md)
- [688292.CN](https://longbridge.com/en/quote/688292.CN.md)

## Related News & Research

- [Tri-Continental Cum.Pref.Shs declares $0.625 dividend](https://longbridge.com/en/news/298050949.md)
- [Mabwell reports Phase Ib/IIa data showing anti-ST2 antibody 9MW1911 cuts COPD exacerbations up to 81%](https://longbridge.com/en/news/298042457.md)
- [Baidu completes dual-primary listing in Hong Kong and Nasdaq](https://longbridge.com/en/news/297602235.md)
- [TenX Protocols Inc. reports Q1 results](https://longbridge.com/en/news/297560433.md)
- [Baidu shifts Hong Kong listing to dual-primary from secondary listing](https://longbridge.com/en/news/297571992.md)

---
> **Disclaimer: This article is for reference only and does not constitute any investment advice.**