Legal name: Suzhou Tongyuan Soft Control Technology Co., Ltd.
Tongyuan Soft Control Deep Dive: MWORKS (Syslab/Sysplorer) as a Domestic MATLAB/Simulink Alternative, with MWORKS AI and the MOGICK Engineering Agent Advancing Home-Grown Physical AI
Tongyuan Soft Control's MWORKS (Syslab scientific computing and Sysplorer Modelica modeling) is a domestic MATLAB/Simulink alternative serving aerospace and nuclear digital twins; in 2026 it launched MWORKS AI and the MOGICK engineering agent and entered IPO tutoring for the Sci-Tech board.
Executive Summary
Tongyuan Soft Control is the only company on this list that writes self-reliance into its technical definition. Its core asset is the Modelica compiler and solver kernel inside the MWORKS platform, which the company describes as one of six commercial Modelica engines worldwide and the only system modelling and simulation environment in the Asia-Pacific region recognised by the Modelica Association. On 7 August 2026 it signed a tutoring agreement with CITIC Securities, and on 14 August 2026 it completed filing with the Jiangsu securities regulator, formally beginning the process of listing on the STAR Market.
This report makes five arguments.
First, Tongyuan physical AI is a three-layer structure and the three layers mature in descending order. The base is the MWORKS platform, covering scientific computing, system modelling, architecture design, data management and digital twins. The middle layer is the Mogick systems engineering agent, with engineering perception, task planning, process orchestration, tool invocation and memory management. The upper layer is the embodied intelligence development environment, which appears publicly under two names, Phyzoic and MWORKS EI. The base has twenty years of engineering validation, the middle layer launched formally in 2026 and remains in beta, and the upper layer has only just been pre-released.
Second, its scarcity comes from the compiler and solver kernel rather than the interface. The barrier in system simulation software is not graphical modelling but solving tens of thousands of equations efficiently and stably. Tongyuan states the kernel is entirely self-developed with no outsourcing at the base layer, and that it supports Modelica 3.6, version 4.0 of the Modelica Standard Library and FMI 1.0, 2.0 and 3.0. If that holds, its substitution value inside China is structural.
Third, the customer structure determines both revenue quality and the ceiling. The company reports more than 70 percent share in aerospace system simulation and more than 50 percent domestic penetration in nuclear and marine power simulation. Customers include the aerospace contractors, China National Nuclear Corporation and COMAC, with work on the space station, the Chang'e lunar programme, the C919, nuclear power and marine propulsion. Such customers bring high unit prices, strong stickiness and long collection cycles, while growth follows national programme timing.
Fourth, the shareholder structure carries dual endorsement from state and industrial capital. The 2022 A round brought in the China Internet Investment Fund, Zhongding Capital, Shenzhen Capital Group and Oriza Holdings at a valuation of about 2.5 billion renminbi. In May 2025 Huawei Hubble entered by acquiring existing shares at a valuation of about 4 billion renminbi, the highest cost basis among institutions. The company converted to a joint-stock company in 2026. Founder Chen Liping holds 28.47 percent directly, and management plus employee platforms control more than 50 percent.
Fifth, the listing is at an early tutoring stage and no financial statements are public. Registered capital is 360 million renminbi, the tutoring sponsor is CITIC Securities, the law firm is Beijing Guantao and the auditor is Zhongxinghua. Market projections about raising 1.5 to 2 billion renminbi and listing in the second half of 2027 are media calculations rather than company disclosure.
Evidence grading. Registration and tutoring filings are primary and verifiable. Product capability and engineering cases are company statements. Market share, user counts and gross margin are company or media claims without third-party audit. Fundraising size and valuation are projections.
Company Overview
Suzhou Tongyuan Soft Control Technology Co., Ltd. was founded on 30 March 2007, is registered at Building D1, 333 Songbei Road in the Suzhou Industrial Park, has registered capital of 360 million renminbi, and names Zhou Fanli as legal representative. Its industry classification is software and information technology services. It completed conversion to a joint-stock company in 2026.
The company originates from the team of professor Chen Liping at Huazhong University of Science and Technology. Chen holds a doctorate in engineering, is a second-tier professor, serves as deputy director of the national intelligent design and numerical control technology innovation centre at the Ministry of Science and Technology, was selected for the Ministry of Education new century talent programme in 2004 and for the national high-level talent programme in 2022. Zhou Fanli also holds a doctorate in engineering, studied under Chen, led development of the multi-domain unified modelling and simulation platform from 2001, co-founded the company in 2007, and serves as chairman and general manager. He is a member of the international Modelica Association, an expert member of the China manned space engineering digitalisation group, and chair of the working committee of the openSCS open-source community.
On qualifications, the company holds national high-tech enterprise status, specialised and sophisticated small giant status, and the national industrial software collaborative research platform designation. In 2024 it led the formation of the Suzhou scientific engineering intelligent computing simulation industrial software innovation consortium, and in December 2025 it was named among Suzhou ten AI tigers.
Shareholding Structure
- Chen Liping, 28.47 percent: controlling shareholder and actual controller
- First employee holding platform, 13.72 percent: centred on chairman Zhou Fanli
- Second employee holding platform, 9.08 percent: research staff
- Zhou Fanli, 7.38 percent: chairman and general manager, direct holding
- Ding Jianwan, 6.92 percent: core research deputy general manager
- Zhongding fund, 5.62 percent: entered in the 2022 A round
- Shenzhen Capital military fund, 5.38 percent
- Gong Xiong, 4.12 percent: early marketing lead
- Shenzhen Capital Group, 3.25 percent
- Huawei Hubble, 2.86 percent: acquired existing shares in May 2025
- China Internet Investment Fund, 2.15 percent: entered in the 2022 A round
- Oriza Suzhou fund, 1.68 percent
- China Resources Trust, 1.12 percent
Founders, management and employee platforms together control more than 50 percent, so control is stable. One detail matters: Huawei Hubble acquired existing shares rather than subscribing new ones, so the transaction changed the shareholder register and created industrial alignment without directly funding research.
Three Common Misreadings
Misreading one: treating Tongyuan as a domestic MATLAB. The core of MATLAB is a scientific computing language and a toolbox ecosystem, while the core of Tongyuan is Modelica system modelling and solving. Its Syslab provides a Julia-based scientific computing environment that covers part of the MATLAB use case, but the technical centre of gravity is system-level modelling and simulation through Sysplorer and Sysblock, which is closer to the Simulink side of MATLAB and Simulink.
Misreading two: reading the Asia-Pacific Modelica recognition as global leadership. The official position is the only officially recognised system modelling and simulation environment in Asia-Pacific and one of six commercial Modelica engines worldwide. That places it among a small group without establishing leadership in performance or ecosystem. Real leadership would show up in solving scale, stability and the richness of industry model libraries.
Misreading three: treating tutoring filing as an imminent listing. The 14 August 2026 filing begins tutoring only. Acceptance of tutoring, application acceptance, exchange inquiries, listing committee review, registration and issuance all follow, and any stage can slip.
Technology
The Six Components of MWORKS
MWORKS is not a single program. It combines six environments plus model libraries. Syslab is the Julia-based scientific computing environment. Sysblock handles block diagram and state machine modelling and simulation, which the company describes as its own capability built by extending Modelica semantics. Sysplorer is the Modelica-based system modelling and simulation environment and the core of the core. Syslink covers model collaboration and data management. Sysbuilder addresses system architecture design in the SysML direction. Syswin provides the system digital twin environment. Basic, professional and industry model libraries sit beneath them.
The completeness deserves recording. The chain runs from architecture design through modelling, computation, collaboration management to twin operation. Many Chinese vendors build single-point tools; very few build the full chain.
The Compiler and Solver Kernel: Where the Barrier Actually Sits
Modelica is a declarative, acausal modelling language. Users describe equations, and the tool is responsible for sorting, symbolically simplifying and solving them. The difficulty concentrates in three places: symbolic handling of the equation system, reducing tens of thousands of equations to a solvable scale; index reduction for differential-algebraic equations; and numerical stability of stiff systems, including step size control and event handling.
Tongyuan states the compiler and solver engine is entirely self-developed and supports Modelica 3.6, version 4.0 of the standard library and FMI 1.0, 2.0 and 3.0. If those hold, its domestic substitution value is structural, because a kernel of this kind cannot be bypassed with an interface wrapper.
A qualification is required. No public third-party benchmark exists for kernel performance. The usual metrics in this category are solving time for million-equation models, stability on event-dense models, and result agreement with Dymola or SimulationX. Tongyuan publishes none of these comparisons.
Sysblock: A Semantic Extension
Modelica does not natively support block diagram and state machine modelling, which is the territory of Simulink and Stateflow. Tongyuan extended Modelica semantics and built Sysblock to cover both.
The technical meaning is that physical modelling and control logic modelling happen on the same kernel, without exporting between two tools. That is a real improvement for system-level simulation, because a realistic equipment model contains both continuous dynamics in mechanics, thermal and electrical domains and discrete controller logic.
Mogick: The Systems Engineering Agent
Mogick is the engineering agent launched in 2026, positioned to turn engineer intent into executable engineering tasks. Five capabilities are published: engineering perception, task planning, process orchestration, tool invocation and memory management. It coordinates tools including SysML, Modelica, CAD and CAE across requirements analysis, design modelling and simulation verification.
One case is shown repeatedly: digital development of a Mars sample return mission, where Mogick supports geometric prototyping, physical modelling, algorithm development and automatic diagnosis and repair of simulation errors.
The most notable capability is that last one. Simulation errors are among the most time-consuming problems in engineering software, since messages often give only a line and an equation index. If an agent can read the error, locate the model fault and propose a fix, the value is direct. Reliability depends heavily on accumulated data and rules, and the product is in beta, with all features free during the period.
The Embodied Intelligence Environment: Two Names
This is the newest block in 2026 and the most confusingly named. At the China International Industry Fair in October 2026 the company showed Phyzoic, while China Daily and the Suzhou industry bureau refer to the same capability as MWORKS EI.
Public descriptions cover high-fidelity three-dimensional physics simulation, configuration of the product body, work scenario and sensors, kinematics and dynamics simulation, reinforcement and imitation learning, model testing and inference deployment on real hardware. Cases include forklift simulation and training, a wheeled dual-arm robot moving parts, and humanoid imitation learning.
Technically it follows a route similar to the NVIDIA Isaac ecosystem of high-fidelity simulation plus reinforcement learning plus sim-to-real. The difference is that it lives on MWORKS and connects to system modelling in Sysplorer and to control design. That is the domestic differentiator: most embodied intelligence simulators handle only robot and scene, while Tongyuan can combine robot, equipment system and control system.
Layering Credibility
Verifiable: registration and tutoring filings; the six-component product structure; Modelica Association recognition, supported by the public record of the association president visiting; the existence of the openSCS community; and the industry fair exhibits.
Partially verifiable: the claim that the kernel is entirely self-developed, which is a company statement without third-party code audit; the claim of being one of six commercial engines, since the association publishes no list; engineering cases without public customer endorsement; and user and ecosystem counts, which are company figures.
Not verifiable: the 70 and 50 percent market share figures, which have no third-party source; gross margin above 75 percent, absent public statements; kernel performance and industry library accuracy; and Mogick task success rates.
Product and Commercial Model
Revenue comes from three blocks that the company calls three business segments.
First, standardised simulation software products, sold as platforms in Sysbuilder, Sysplorer and Syslink plus 37 industry-specific model libraries, licensed by seat or module. This is the highest-margin block, with company and media reporting software gross margin above 75 percent.
Second, MBSE digital engineering services, providing lifecycle digitalisation consulting, custom model library development and co-simulation integration for state-owned enterprises including the aerospace contractors, China National Nuclear Corporation and COMAC. This block is project-based, with revenue recognised on milestones and cash flow following project progress.
Third, industry-level digital twin solutions, building virtual test platforms for large aircraft, launch vehicles, nuclear plants and new energy vehicles to replace overseas simulation software across full-model testing.
The key commercial constraint is the customer mix. Aerospace, nuclear and marine procurement involves long cycles, long decision chains and heavy dependence on national programme timing. Three consequences follow: project revenue is hard to predict, customer concentration is high, and once a supplier enters a model programme, switching cost is very high because a model programme often runs longer than a decade.
Ecosystem figures deserve separate recording. Company statements cite more than 350,000 users, more than 1,000 universities, more than 1,300 enterprises and more than 5,000 developers. More than 100 universities offer over 2,000 MWORKS-related courses, and more than 20 books have been published. The education investment logic mirrors MathWorks: cultivate future engineers.
Open source positioning. The company leads the openSCS community for open scientific computing and system modelling, with Zhou Fanli as working committee chair, and three core projects entered the Suzhou Zijin open-source incubation programme. The strategic purpose is to establish a de facto standard through open source while keeping commercial value in engineering services and industry libraries.
Competitive Landscape
Against MathWorks with MATLAB and Simulink, the comparison is the most direct. Simulink is the global standard in system modelling and simulation. Tongyuan differentiates through the open Modelica standard route, which avoids locking into one vendor's modelling language, and through domestic compliance, meaning data does not leave the country and the stack is adapted to domestic IT systems. The technical gap lies mainly in the breadth of the toolbox ecosystem and the maturity of control design.
Against Siemens with Simcenter Amesim, the contest is between two poles of system simulation. Amesim has deep accumulated capability in hydraulic, pneumatic and thermal physical modelling, while Tongyuan industry libraries fit domestic programme requirements more closely in aerospace and nuclear. Competition turns on who supplies the more complete industry library.
Against Dassault with Dymola, this is a direct contest inside the Modelica camp. Dymola is the benchmark among commercial Modelica engines and the comparison Tongyuan must win technically. Tongyuan advantages are local service and domestic qualifications; disadvantages are international ecosystem and third-party library richness.
Against domestic CAE vendors including Suocen, ZWSoft and Anshi Pacific, the relationship is more complementary. Tongyuan works at system level, using zero and one-dimensional models, while most domestic CAE vendors work at component level in three dimensions. The realistic collaboration is system plus component simulation, for example a full-machine model in Sysplorer checked against high-fidelity CFD or FEA on key parts. Tongyuan already has a joint CAD plus system simulation offering with ZWSoft.
Against NVIDIA in embodied intelligence, potential competition and dependency coexist. Phyzoic overlaps Isaac Sim in capability, but Tongyuan positions itself as a domestic substitute integrated with systems engineering. It does not depend on NVIDIA compute, and it does not receive NVIDIA ecosystem resources.
Against US export controls, the situation is both risk and opportunity. Restrictions on MATLAB use at some Chinese universities directly pushed domestic substitution demand, and Tongyuan is the direct beneficiary of that logic.
Financing and Capital Structure
The financing history divides cleanly into three stages.
The 2022 A round brought the China Internet Investment Fund, Zhongding Capital, Shenzhen Capital Group and Oriza Holdings at a valuation of about 2.5 billion renminbi. Those shareholders remain on the register: the Zhongding fund at 5.62 percent, the Shenzhen Capital military fund at 5.38 percent, Shenzhen Capital Group at 3.25 percent, the China Internet Investment Fund at 2.15 percent and the Oriza fund at 1.68 percent.
The May 2025 strategic round brought Huawei Hubble in through a purchase of existing shares for 2.86 percent, at a valuation of about 4 billion renminbi. This carries the highest cost basis among institutions, estimated at 9.5 to 10.2 renminbi per unit of registered capital against roughly 7 renminbi in the A round. The logic is industrial alignment, integrating MWORKS into Huawei cloud, intelligent vehicle and factory digital twin ecosystems.
Conversion and tutoring in 2026. The joint-stock conversion completed in 2026, the tutoring agreement with CITIC Securities was signed on 7 August, and filing with the Jiangsu regulator followed on 14 August, with Beijing Guantao as counsel and Zhongxinghua as auditor.
On fundraising and valuation. No company-confirmed figure for the offering exists. Media projections suggest 1.5 to 2 billion renminbi for kernel iteration, an aerospace digital research centre, capacity in new energy and nuclear solutions, and a national sales and service network. Such figures are estimates and should not be cited as fact.
On revenue transparency. No financial statements are public, so revenue, profit and customer concentration have no primary data. The only usable reference is the company and media statement that software gross margin has long been above 75 percent.
Public Bid Awards and Employee Scale (October 2026)
In October 2026 Tongyuan disclosed several awards and shortlists on public procurement platforms: on October 21 the China Ship Research and Design Center published a shortlist for an adaptability digital prototype model configuration management tool and two other equipment types, with a results notice on October 27 awarding the contract at 970,000 renminbi; on October 27 AVIC Civil Aircraft published a shortlist for a hardware-in-the-loop system test software platform where Tongyuan ranked third; around October 26 to 28 the China United Gas Turbine Technology published a procurement result for a heavy gas turbine digital prototype three-dimensional visualisation module; and on October 22 Yantai University accepted a mooring positioning system simulation analysis software project. Business registry information separately discloses an employee scale of about 200 to 299, 217 insured persons, registered capital of 360 million renminbi and 617 cumulative bidding participations. The awarded amounts, all from public procurement notices, are individually small and reflect a customer base of central state-owned enterprises and research institutes served through project-based delivery.
DISCREPANCIES
1. The tutoring period is described two ways. (Source grade: OFFICIAL versus MEDIA) The tutoring filing states a period from August to October 2026, following the agreement signed with CITIC Securities on 7 August. Media analysis projects a conventional six to eight month cycle running to tutoring acceptance in March 2027. The first is the filed period, the second a projection covering acceptance and application. This report uses the filed period and flags the other as a projection.
2. The embodied intelligence environment carries two names. (Source grade: OFFICIAL) The October 2026 industry fair and company channels use Phyzoic, while China Daily and the Suzhou industry bureau use MWORKS EI, and other material describes MWORKS EI as pre-released. The described capabilities overlap heavily, covering body, perception and environment modelling, parallel simulation training, synthetic data and sim-to-real. This report treats them as two names for one product line and advises readers to check the latest official naming.
3. Market share figures lack a third-party source. (Source grade: OFFICIAL and MEDIA) The claims of more than 70 percent share in aerospace system simulation and more than 50 percent domestic penetration in nuclear and marine power appear in company and media material without third-party data. This report cites them as company statements.
4. Gross margin above 75 percent has no statement support. (Source grade: MEDIA) The figure comes from media description of profitability and no statements are public. This report labels it company and media language.
5. Financing rounds are described inconsistently. (Source grade: MEDIA) Some material says a 2022 A round and an April 2025 A-plus round, while other material says Huawei Hubble acquired existing shares in May 2025. This report follows the shareholding record in the registration and tutoring filings: a 2022 A round subscription and a May 2025 transfer of existing shares, and does not adopt the A-plus phrasing.
6. User scale changes with time. (Source grade: OFFICIAL) The October 2026 industry fair figure is more than 350,000 users, more than 1,000 universities, more than 1,300 enterprises and more than 5,000 developers, while earlier material cites more than 250,000 users. This report uses the October 2026 figures.
7. The list of six commercial Modelica engines is not public. (Source grade: OFFICIAL) The company states its engine is one of six commercial engines worldwide, but the Modelica Association publishes no such list, so the claim cannot be checked independently. This report retains the statement and labels it unverifiable.
Risks and Diligence Gaps
Listing uncertainty is the first risk. Only tutoring filing is complete. Many stages follow, and financial normalisation plus verification of the shareholding history, given multiple capital increases and share transfers, can extend the tutoring period.
Customer concentration and national programme timing. Orders in aerospace, nuclear and marine power track national programme approvals, so annual revenue is volatile. If one programme winds down before the next is approved, a revenue gap appears.
Revenue scale is undisclosed, so growth cannot be quantified. This is the largest obstacle to investment judgment. For reference, only a handful of listed Chinese industrial software companies exceed 1 billion renminbi of revenue, and Tongyuan scale will only be visible in the prospectus.
Agent and embodied intelligence commercialisation is early. Mogick is in beta with all features free, and the embodied environment has only just been pre-released. Both are cost centres in the near term.
Verifying kernel independence is costly. The claim of complete self-development is hard to verify. The real tests are code audit and third-party benchmarking, neither of which exists.
The ecosystem depends on an open-source community. Whether openSCS becomes a de facto standard depends on activity, contributor scale and project maturity, all requiring sustained investment.
Diligence gap list, ordered by importance. First, revenue, gross margin, net profit and customer concentration for the last three years, available once the prospectus is published. Second, the split between software and services revenue. Third, names and contract values of the top five customers. Fourth, third-party benchmarks for the kernel, including million-equation solving time and result agreement with Dymola. Fifth, the number, coverage and validation method of industry model libraries. Sixth, the paid conversion path for Mogick once the free beta ends. Seventh, trial customers for the embodied intelligence environment. Eighth, the specific joint projects and amounts following the Huawei Hubble investment.
Composite Assessment
techScore 8.0. A self-developed Modelica compiler and solver kernel is a scarce domestic asset, the six MWORKS components cover the full system simulation chain, and the Sysblock semantic extension solves the real problem of modelling physics and control on one kernel. The deduction reflects the absence of third-party kernel benchmarks and undisclosed accuracy of industry libraries.
bizScore 7.0. Reported share above 70 percent in aerospace system simulation, software gross margin above 75 percent, and customers among state-owned enterprises and national programmes describe good commercial quality. The deduction reflects undisclosed revenue scale, volatile project revenue and high customer concentration.
fundScore 7.5. A 2022 A round at a 2.5 billion renminbi valuation, Huawei Hubble entering in 2025 at about 4 billion, and the STAR Market tutoring started in 2026 form a clear financing path with both state and industrial capital. The deduction reflects the unlisted status and undefined offering size.
productScore 7.5. The chain from architecture design to twin operation is complete, and Mogick plus the embodied environment add the interface between AI and the physical world. The deduction reflects new products at an early stage and an ecosystem toolchain less rich than international leaders.
marketScore 7.5. Policy-driven demand for domestic substitution, qualification barriers in national programmes and an education base of 350,000 users form a solid foundation. The deduction reflects a market bounded by domestic high-end equipment research spending.
References
- Jiangsu securities regulator tutoring filing dated 14 August 2026: founded 30 March 2007, registered capital 360 million renminbi, legal representative Zhou Fanli, registered at Building D1, 333 Songbei Road in the Suzhou Industrial Park, controlling shareholder Chen Liping at 28.47 percent, industry code I65, tutoring sponsor CITIC Securities, counsel Beijing Guantao and auditor Zhongxinghua
- Tutoring agreement signed with CITIC Securities on 7 August 2026 with a filed tutoring period from August to October 2026
- Shareholding per registration and tutoring filings: Chen Liping 28.47 percent, first employee platform 13.72 percent, second employee platform 9.08 percent, Zhou Fanli 7.38 percent, Ding Jianwan 6.92 percent, Zhongding fund 5.62 percent, Shenzhen Capital military fund 5.38 percent, Gong Xiong 4.12 percent, Shenzhen Capital Group 3.25 percent, Huawei Hubble 2.86 percent, China Internet Investment Fund 2.15 percent, Oriza Suzhou 1.68 percent, China Resources Trust 1.12 percent
- Financing history: the 2022 A round with the China Internet Investment Fund, Zhongding Capital, Shenzhen Capital Group and Oriza at a valuation of about 2.5 billion renminbi; Huawei Hubble acquiring existing shares in May 2025 at a valuation of about 4 billion renminbi for 2.86 percent; joint-stock conversion completed in 2026
- MWORKS product structure: Syslab for Julia-based scientific computing, Sysblock for block diagrams and state machines, Sysplorer for Modelica system modelling, Syslink for model collaboration and data management, Sysbuilder for architecture design, Syswin for system digital twins, plus basic, professional and industry model libraries
- Modelica capability: Sysplorer is the only officially recognised system modelling and simulation environment in the Asia-Pacific region and supports Modelica 3.6, version 4.0 of the standard library and FMI 1.0, 2.0 and 3.0; the self-developed compiler and solver engine is described as one of six commercial engines worldwide
- Record of the Modelica Association visit on 25 September 2026, when association president Dirk Zimmer and model library project lead Francesco Casella visited the Suzhou headquarters to discuss MWORKS development on Modelica and China's contribution to the international ecosystem
- Mogick capabilities: engineering perception, task planning, process orchestration, tool invocation and memory management, coordinating SysML, Modelica, CAD and CAE; the demonstrated case is digital development of a Mars sample return mission covering geometric prototyping, physical modelling, algorithm development and automatic diagnosis and repair of simulation errors; all features free during beta
- Embodied intelligence environment under the names Phyzoic and MWORKS EI: high-fidelity three-dimensional physics simulation, configuration of body, scene and sensors, kinematics and dynamics simulation, reinforcement and imitation learning, model testing and deployment; cases include forklift simulation and training, a wheeled dual-arm robot handling parts, and humanoid imitation learning
- Ecosystem scale as of the October 2026 China International Industry Fair: more than 350,000 users, more than 1,000 universities, more than 1,300 enterprises and more than 5,000 developers; more than 100 universities offering over 2,000 courses; more than 20 books published; three openSCS core projects entering the Suzhou Zijin open-source incubation programme
- Engineering practice and customers: aerospace contractors, China National Nuclear Corporation and COMAC; participation in the space station, the Chang'e programme, the C919, nuclear power and marine propulsion; reported share above 70 percent in aerospace system simulation and above 50 percent domestic penetration in nuclear and marine power, as company statements
- Qualifications and recognition: national high-tech enterprise, specialised and sophisticated small giant, national industrial software collaborative research platform, leadership of the Suzhou innovation consortium formed in 2024, and selection among Suzhou ten AI tigers in December 2025
- Public procurement platform result notices and business registry information, October 2026: the China Ship Research and Design Center awarded a contract at 970,000 renminbi for an adaptability digital prototype model configuration management tool and two other equipment types (results notice October 27); AVIC Civil Aircraft shortlisted Tongyuan third for a hardware-in-the-loop system test software platform (October 27); the China United Gas Turbine Technology published a procurement result for a heavy gas turbine digital prototype three-dimensional visualisation module (October 26 to 28); Yantai University accepted a mooring positioning system simulation analysis software project (October 22); business registry discloses about 200 to 299 employees, 217 insured persons, registered capital of 360 million renminbi and 617 cumulative bidding participations (amounts and employee scale from public notices and registry, not independently audited)
This report is compiled from public sources for informational purposes only and does not constitute investment advice. Data is current as of 2 November 2026; please refer to company announcements and periodic reports.
Related papers
Related tools
Related cases
- DrivAerML High-Fidelity Road-Car Aerodynamics Dataset and Surrogates
- AirfRANS High-Fidelity Airfoil Surrogate Benchmark Case
- Physics-Informed CNN for Battery Pack Temperature Fields
- IASC-ASCE Structural Health Monitoring Benchmark and Physics-Informed Damage Identification
- OpenFOAM Industrial Multiphysics Pipeline: From Meshing to Data Supply
Related reports66
- 51World Deep Dive: A Listed Digital Twin Company Where 51Sim Becomes the First Growth Engine While 51Earth Nearly Vanishes, and Two Irreconcilable Financing Naming Conventions Must Both Be Kept
- AMI Labs: A USD 1.03B Seed Bet on World Models, LeCun's Non-Generative Path and Its Blank Scorecard
- Anshi Pacific (PERA) Deep Dive: A Beijing Domestic-CAE Leader Founded in 2003 Whose PERA SIM General Simulation, AIROM Reduced-Order Models, AgentX Multi-Agent and Jingzhi iGPT/Whale Span from Industrial Simulation to Physics AI
- AspenTech Deep Dive: The Process-Simulation Incumbent Acquired by Emerson and Its Hybrid Industrial-AI Route
- Autodesk Deep Dive: Neural CAD Foundation Models and a $200M World Labs Bet on Physical-World AI, with NAVASTO Adding CFD Surrogate Capability
- Bentley Systems Deep Dive: iTwin Digital Twin and Generative AI Empower Full-Lifecycle Infrastructure Engineering
- BeyondMath Deep Dive: A Foundation Physics AI on the Neural-Operator Route, a $19M Consortium Budget, and Three Misread Facts
- Cadence Deep Dive: System-Level Multiphysics EDA and the Millennium M2000 Supercomputer Anchor Physical AI, with AuraStack Agentic AI and Fidelity GPU CFD as Key Levers
- COMSOL Deep Dive: Four Decades of Multiphysics Solvers, Entering Physical AI via DNN Surrogate-Model Training and All-Physics GPU Acceleration
- Dassault Systèmes Deep Dive: 3DEXPERIENCE and SIMULIA Virtual Twin Physics Behavior Anchor Industrial AI, with LEO Virtual Companion and GPU Explicit Dynamics as Physical-AI Levers
- DP Technology: From the Gordon Bell Prize to National-Level Unicorn Status and the Door of a Hong Kong Listing, the AI-for-Science Leader's Technical Delivery and Commercial Blank
- Emmi AI: From Linz Spin-out to a EUR 330M Mistral Acquisition in 17 Months
- Fainite Deep Dive: A Zurich Physics-AI Simulation Startup Whose Caltech, ETH Zurich, Cambridge and Google Team Uses Physics-Aware Deep Learning to Cut Simulation from Tens of Hours to Seconds, With Only Venture Kick Grants and No Priced Round
- Fysics AI: The PhysX Founder's Second Act, a Domestic Differentiable Physics Engine and Its Ecosystem Play
- Frontop (301313) Deep Dive: From Digital Exhibition Halls to a Physical-AI Data Foundation - FTE Engine, FunSim Simulation and the Gap Between Positioning and Revenue
- Godela Deep Dive: A YC X25 Physics-AI Simulation Startup Whose Ex-Apple, Google and Intel Team Uses Physics-Constrained Surrogate Models to Cut Simulation from Days to Seconds, at Seed Stage with Pilot-Only Access
- HollyWave (688682) Deep Dive: RDSim AI Electromagnetic Simulation and Engineering Agents Close the Simulation-Measurement-Product Loop
- IAT (Alt) Deep Dive: China's First Listed Independent Automotive Design Company Pivots to AI-native R&D with the Yuanfang, Yufeng and Taiyi 3.0 Product System and an NVIDIA Omniverse Partnership, Reporting RMB 975M Revenue and a Widened RMB 250M Net Loss in 2025
- Inductiva Deep Dive: A Closed Case in Open-Solver Orchestration, Eight Premises That Must Be Corrected, and Where Co-Founder and CTO Hugo Penedones Went Next
- Intesim (INTESIM) Deep Dive: A Home-grown Chinese CAE Leader Founded in 2009 with Self-developed Solvers for Structure, Fluid, Electromagnetics, Thermal and Acoustics and Multiphysics Coupling, a ~RMB 100M B Round from CNII and Shenzhen Capital, and 2025 Revenue and Bookings Both above RMB 100M
- iSoftStone (301236) Deep Dive: A ChiNext-Listed IT Services Giant Whose TianShu Platform Builds a Three-Layer Physical-AI Stack with Self-Developed TwinHub Digital Twin, FSim Industrial Simulation, CAE Multiphysics and Phyxis Embodied Control
- Jua Deep Dive: The Earth Physics Transformer as a Large Physics Model for Energy Trading
- Keysight (with ESI Group) Deep Dive: Virtual Prototyping and Hybrid AI Bridge Measurement and Simulation, with Natural-Language Assistants in EDA
- Lightwheel Deep Dive: A Physical AI Data and Evaluation Infrastructure Startup Founded by a Former NVIDIA Simulation Lead, Seven Rounds Totalling Roughly CNY 2.65 Billion at a Valuation Above USD 2 Billion, Seats on Both the Newton and EgoVerse International Standards Committees, With Two Incompatible Accounts of Q1 Orders
- Luminary Cloud: The Physics AI Model Factory Backed by NVentures
- MathWorks Deep Dive: MATLAB/Simulink and Simscape Multiphysics Power Physical AI, with R2026a Copilots and an Agentic Toolkit for Edge Intelligence
- Maniformer (Mifeng) Deep Dive: An Agibot-Spun Physical-AI Data Infrastructure Firm Whose Bodyless MEgo Capture Hardware Is in Mass Production and Which Closed Three Rounds in Half a Year Led by China Telecom
- Mistral AI: How Physics AI Moved From Language Models to EUV Lithography and Nuclear Power
- Monolith AI: The British Engineering-AI Firm That Cut Physical Test Counts Instead of Accelerating Simulation, Now Inside CoreWeave
- Monumo Deep Dive: A Cambridge Deeptech Company Redesigning Electric Motors with the Anser AI Engine, Tens of Millions of Multiphysics Simulations per Day, GBP 10.5M Raised in Total, and a Series A Being Raised
- NAVASTO: Moving Solver Insight Onto the Designer's Desk, and How Autodesk Bought That Entry Point
- Navier: Handing CFD Workflows to Agents, a Seed-Stage Bet on AI for Hardware Engineering
- Neural Concept: An EPFL Spin-out Betting on Neural Surrogates, with $138M Raised
- NP Co. (NP Company) Deep Dive: A 2025-founded Paris Inria-born AI-native simulation startup that raised a EUR 6M pre-seed led by Partech with Mistral AI co-founders as angels, claiming a pre-trained physics foundation model at 1,000x speedup — most claims unverified
- NVIDIA Deep Dive: Building the Full-Stack Physical-AI Infrastructure with PhysicsNeMo, Warp, Cosmos World Foundation Models and Omniverse
- Physical Intelligence Deep Dive: A $2.1B Bet on the GPT Moment for Robotics, With pi0.7 Self-Reporting 60-80 Percent Success on Unseen Tasks and Neither Pricing nor Revenue Disclosed for the Pi API
- Physical Superintelligence Deep Dive: A $58M Seed Bet on Virtual Physicists, With Emmy Targeting AI Data Center Multiphysics Optimization as Its First Commercial Proof Point
- PhysicsX: Pretrained Geometry and Physics for Industrial Engineering at a $2.4B Mark
- Prometheus: A $41B Mark, $18.2B Raised, and Not Yet a Product
- PTC Deep Dive: Creo/Onshape and the ThingWorx Digital Thread Power Physical AI, with NVIDIA Isaac Sim Bridging Robotics Design to Simulation
- Qianjue Technology Deep Dive: How a Body-Agnostic Embodied Brain Pairing the Predictive World Model Polibrain with Brain-Inspired Partitioning and Polynomial Representation Reached a Hundred Thousand Cleaning and Service Robots
- Quanscient Deep Dive: Cloud-Native and Quantum-Powered Multiphysics Simulation as the Data Engine for AI-Native Hardware Engineering
- Rescale Deep Dive: An HPC Operating System and Desktop Engineering Workbench in the Cloud, a $115M Series D, and Five Misread Facts
- Schrödinger Deep Dive: Physics-Based FEP+ Simulation and Machine Learning Power Molecular Discovery, with the 2026 Bunsen Agentic Co-Scientist
- Secondmind Deep Dive: A Cambridge Engineering-AI Firm Whose Bayesian Active Learning Cuts CAE Simulation and Test-Bench Runs by up to 90%, with Roughly USD 73M Raised and Mazda Both Customer and Investor
- Shichuang Technology Deep Dive: SupreCAST Cloud-Native Material-Processing Simulation and Industrial Agents Advance the CAE+AI Strategy
- ShonCloud (Shunyun) Deep Dive: Meshless Particle-Method CFD for EV and Nuclear Fluid Multiphysics
- Shuzhi Maijin Deep Dive: A Shenzhen Fully Self-Developed SNN-plus-LBM CFD Startup Claiming Second-Level Results on Hundreds-of-Millions of Grids on Consumer Hardware, With a Tens-of-Millions-RMB Angel Round From a Ministry-of-Finance-Affiliated Fund and an FPGA Migration Plan
- Siemens (incl. Altair) Deep Dive: The 2025 Altair Acquisition Unites Simcenter and Altair, with PhysicsAI Geometric Deep Learning and PhysicsAI Generate as the Physical-AI Core
- Silurian Deep Dive: A Generative Forecasting Transformer Earth-System Foundation Model for Weather Risk on Energy and Infrastructure Assets
- SimScale Deep Dive: A Cloud CAE Platform Built on Core-Hour Subscriptions, EUR 52M of Equity Kept Separate from EUR 25M of Debt, and Five Misread Facts
- Skild AI Deep Dive: A General Robot Foundation Model Incubated at CMU's Robotics Institute, With a USD 1.4B Series C at a USD 14B Valuation, a Conflicting Series B Size, and an ARR Figure That Is a Run-Rate Rather Than Revenue
- Songying Deep Dive: A Huawei-Cloud-Alumni Startup Building a Domestic Physical AI Operating System, With ORCA OS Positioning Against NVIDIA Omniverse, Three Incompatible Financing Narratives, and a Verifiable Third-Party Anchor in a CNY 20 Million Minority Investment
- Sudo Tech Deep Dive: Zero-Real-Data Simulation Training, the Fully In-House Sudo R1 Stack, and an Atomic-Skill Commercialisation Strategy from a Startup Valued Above CNY 20 Billion in 17 Months
- Suochen Technology: The Pain of Becoming a Physics-AI Platform from a CAE Vendor
- Synopsys (Ansys) Deep Dive: The 2025 Ansys Acquisition Turns the EDA Leader into a Silicon-to-Systems Engineering Powerhouse, with Ansys SimAI and the NVIDIA Physical-AI Partnership as the Core Physics-AI Levers
- TenFong Deep Dive: An Academician-Led Chinese Industrial Simulation Software Vendor Whose Official Timeline and Database Records Use Incompatible Round Names, With an Official Financing PDF and an Investment Database Both on File
- Tianfu (Nanjing Tianfu Software) Deep Dive: A 2011-founded Chinese Intelligent-CAE Veteran That Closed a Several-Hundred-Million-RMB Series C, Pivoted to an Industrial World Model in 2024, With Unreconcilable Financing and Valuation Records and No Disclosed Revenue
- Vinci AI: Turning Physics Simulation from a Checkpoint into a Continuous Constraint
- Westlake Robotics Deep Dive: Westlake University's First Robotics Spin-out Builds a Dual-Pretrain Brain-and-Cerebellum Architecture with WR1 Whole-Body Foundation Model and GAE Behaviour Expert
- World Labs: Fei-Fei Li's Spatial-Intelligence Unicorn, Acquired by AMD for $8.2B
- XVERSE Deep Dive: A Simulation-World Route to Physical AI Built on the XVERSE Model Family, an Edge-Cloud 3D Engine and the Physics-Based MotionGen Model
- SimWE: From In-House Solvers to a Physical-AI Engine, the Domestic CAE Leader and What Comes Next
- Yunjing Zhifang Deep Dive: A Peking-University-Incubated Cloud-Native GPU CAE Company with the Simversus Platform and the YJ-DORGP Small-Sample Intelligent Solver, Reporting 275 Percent Order Growth, 160 Percent Revenue Growth and Breakeven in 2025
- Zhongke Diwiji (Five Ages) Deep Dive: A CAS-Institute-of-Automation and Tsinghua Spin-out Whose FAM Embodied Foundation Model Uses Action Optical Flow for Cross-Embodiment Generalisation, Reaching 97% Success from 3 to 5 Demos, with EraBrain and Jiling Robots Serving Sinopec, Leapmotor and Midea
- ZWSOFT (688083) Deep Dive: Overdrive Kernel and ZW3D Wukong 2027 Underpin an Integrated CAD/CAE/CAM Stack, with a CAx Agent Platform as the Practical Route to Engineering AI
This report is compiled from public sources. It is for reference only and does not constitute investment advice. Data is current as of the publication date; please refer to the company’s official filings.