Car manufacturers are competing in intelligence; is this the opportunity for domestic intelligent platforms?

```

In the second half of the automotive industry's rapid shift towards intelligence, many players have set their sights on the smart cockpit track.

In the past few years, Qualcomm’s cockpit SoCs represented by Snapdragon 8155 and 8295 once captured more than 70%—even up to 80%—of the domestic passenger car market.

With the competition intensifying in the main price range of 100,000 to 200,000 RMB, car companies’ anxiety about cost control and supply chain security has reached its peak, opening a historic window for large-scale deployment of domestic chips. However, domestic chips need more than just hardware to seize market share.

On April 25, during the Beijing Auto Show, Runxin Micro Technology, which had just completed nearly 400 million RMB in B+ round financing, officially launched its new “Jizuo Strategy” based on domestic chips. This six-year-old company is attempting to convey a signal: there is a possibility to break through the bottleneck of hardware-software integration for large-scale adoption of domestic chips.

However, currently, Runxin Micro’s clients are mainly commercial vehicles and mid-to-low-end passenger cars. This domestic player still has some distance from shaking up Qualcomm’s main market.

The Dilemma of Domestic Cockpit Chips

According to data from the Gasgoo Auto Research Institute, in January and February 2026, Qualcomm held 72.1% of the market share for smart cockpit chip installations in domestic passenger cars, with shipments reaching 1.014 million units. This means that nearly three out of every four smart cars have cockpit chips from a single overseas manufacturer.

This highly concentrated supply chain configuration is not without concerns. During the global chip shortage in 2021, several automakers were forced to cut production or delay deliveries due to tight cockpit chip supplies, leaving a deep impression on the industry. From the perspective of supply chain resilience, a single supplier holding such a high share is, in itself, a risk exposure.

Domestic players have made considerable progress in the past two years. According to Zosi Auto Research, the domestic localization rate of cockpit domain controller SoCs reached about 18% between January and November 2025, whereas before 2022, that figure was less than 3%.

However, the gap between “useable” and “good to use” for domestic chips lies in a deep divide of hardware-software collaboration. Chip manufacturers excel at hardware design, but lack deep optimization for automotive-grade operating systems and AI middleware; automakers control vehicle definition and user scenarios, but usually do not dive into adaptation layers of chip firmware and domain controller architecture.

Zhang Mingjun, co-founder and CTO of Runxin Micro, explained to Wallstreetcn on April 25 that only when domestic chips begin to be widely used can issues be discovered and prompt continuous iteration and optimization—a path that was missing in the past.

Traditionally, this bridge was occupied by international Tier 1s or Qualcomm’s own software solutions. For domestic chips to achieve large-scale deployment, deep coupling between chip firmware, vehicle-side OS, and domain controller architecture must be solved. This is the direction many companies are exploring and is believed to be the core reason that domestic chips’ penetration rate finds it difficult to quickly break the 10% threshold.

There are many players in this track. Desay SV, Pateo, Neusoft Group and other Tier 1s have accumulated years of experience and established relationships with leading automakers in the cockpit domain field. Chip companies like SemiDrive are also pushing their own SoC software platforms. Comparatively, Runxin Micro differentiates itself by not binding to a single chip but adapting to multiple chip platforms, which also means it must maintain multiple technology stacks, putting considerable pressure on R&D investment.

Meanwhile, the penetration rate of smart cockpits is approaching its ceiling.

Industry data shows the penetration rate of smart cockpits was about 59% in 2022, and surpassed 80% in 2026. As incremental opportunities shrink, the market shifts from expanding the pie to slicing it. It’s getting harder for new entrants to make their way onto this list.

Runxin Micro’s Solution

Runxin Micro announced its “1+4+N” strategy at the Beijing Auto Show, attempting to present its answer.

The “1” refers to a domestic AI smart foundation composed of “Zhixin·Domestic Computing Platform + Zhiwei·AIOS + Zhirun·Edge Model.” The “4” stands for four core tracks: vehicle side, mobile side, AI smart hardware, and embodied intelligence. “N” refers to extending into scenarios like mobility, industry, and home.

Simply put, Runxin Micro doesn’t make chips, but offers an “chip + OS + AI” full-stack adaptation solution, acting as a middleware between domestic chips and automotive-grade applications.

In terms of products, Runxin Micro unveiled two computing platforms.

The C200 is its main production foundation, with computing power of 60K-90K, 90% domestic content, deployed in over ten vehicle models. The X100 targets higher-end markets, integrating cockpit and driving, offering 100K+80T computing power, about 85% domestic content, using Horizon J6E chip, supports highway NOA and automatic parking.

Additionally, the company released the Zhiwei·AIOS Lite intelligent agent operating system built on openvela, and the Rbox-S100 “big + small brain” integrated computing platform for robotics.

Mass production scale is the hard indicator for solution feasibility.

Runxin Micro claims to have achieved production and deployment of its smart cockpit solutions in over one million vehicles within five years, supporting passenger and commercial vehicle models from SAIC, BAIC, Dongfeng, etc., with monthly shipments steadily increasing.

Regarding chip adaptation, Runxin Micro has completed full-stack adaptation for mainstream domestic chips including Unisoc, AutoChips, Horizon, etc. Among these, its cooperation with Unisoc is the deepest: the C200 platform deeply customizes around the A7870 chip, mass-produced and launched with SAIC Maxus wide-body vans and new pickup models, and exported to overseas markets.

This case illustrates that domestic chip solutions can not only succeed domestically, but also go abroad with Chinese vehicles. Zhang Mingjun revealed that previous overseas shipments mainly followed OEM products, but now they are also exploring direct cooperation with overseas companies.

In January 2026, Runxin Micro completed a nearly 400 million RMB B+ round of financing, with investors including Chongqing Changjia Zongheng, Hengxu Capital, Jiangsu Strategic Emerging Industry Fund, all state-owned or industrial capital.

The funds will be used for computing platform upgrades and edge model development.

But there are also direct problems.

First, their market position remains a follower. Their main production clients are commercial vehicles and mid-to-low-end passenger cars from SAIC, BAIC, Dongfeng, etc., with a customer level gap compared to top players.

Second, price pressure persists. The smart cockpit industry generally faces a “annual price reduction” pressure of 4% to 6%. For suppliers with limited revenue scale, profit margins will be further squeezed.

Runxin Micro’s “1+4+N” strategy essentially attempts to fill the gap of “hardware-software integration + experience translation” in the grand narrative of domestic chip substitution. The achievement of one million vehicles shows the solution is feasible, and state-owned financing highlights its strategic value in the independently controllable value chain.

But whether they can move from one million units to larger scale, from mid-to-low-end models to higher-value markets, and from single cockpits to cockpit-driving integration will determine if this company can truly stand firm in the cockpit track, where penetration is peaking and giants abound.

Risk disclaimer and exemption clauseThe market carries risks; investment should be approached with caution. This article does not constitute personal investment advice, nor does it take into consideration the specific investment objectives, financial situation, or needs of any individual user. Users should assess whether any opinions, views, or conclusions in this article are suitable for their particular circumstances. Investing accordingly is at your own risk. ```