What Q.ANT does

Q.ANT GmbH is a Stuttgart-based photonic computing company founded in 2018 as a spin-out from TRUMPF's research laboratories. It designs and manufactures Thin-Film Lithium Niobate (TFLN) photonic processors and quantum sensors that perform mathematical operations natively in the optical domain, where one optical element can replace 100–1,000 transistors for nonlinear functions. The company operates two business units: Native Computing, which commercializes the Native Processing Unit (NPU) and the Native Processing Server (NPS) as PCIe-compatible co-processors for AI training/inference and HPC workloads; and Native Sensing, which sells the Q.M 10 photonic quantum magnetometer (10 Picotesla sensitivity) for medical technology applications including prosthetics and neurology.

The company's commercial product line has progressed from a September 2024 cloud-access NPU demo (MNIST handwriting recognition, 95% accuracy) to the November 2024 launch of the first commercial NPU, the November 2025 unveiling of the second-generation NPU (8 GOPS sustained on nonlinear functions, up to 50x throughput and 6x lower energy vs. Gen 1), and the May 2026 IONOS agreement marking the first commercial cloud customer for the NPS. Production is anchored by a TFLN chip pilot line in Stuttgart operated jointly with IMS CHIPS on refurbished CMOS fabrication infrastructure. Customers to date include two flagship European HPC centers (Leibniz Supercomputing Centre since July 2025; Jülich Supercomputing Centre since November 2025), one commercial cloud anchor (IONOS), and a small set of developer/research collaborators (Daisytuner for PyTorch compilation, NXAI for xLSTM workloads, LMU Munich for diffusion models).

Q.ANT's revenue model combines hardware sales (NPS and Q.M 10), freemium cloud-access evaluation, and government-funded R&D program revenue (BMBF/BMFTR projects including the LichtBriQ consortium). Pricing is not publicly disclosed; commercial contracts are negotiated directly. Following a Series A totaling approximately $80 million across two 2025 closings led by Duquesne Family Office, the company has expanded internationally with a US headquarters in Austin, Texas opened in April 2026, and is recognized as a Sample Vendor in three 2025 Gartner Hype Cycle reports including Data Center Infrastructure Technologies.

Name

Q.ANT

Legal name

Q.ANT GmbH

Website

https://qant.com

Company type

Private

Founded year

2018

Operating status

Operating

Headcount range

51–100 employees

Short description

Q.ANT is a German photonic computing company that develops Thin-Film Lithium Niobate (TFLN) processors and quantum sensors for energy-efficient AI and HPC workloads, serving hyperscalers, cloud providers (IONOS), and European supercomputing centers (LRZ, Jülich).

Ownership category

akta.pro rank

Product category

Photonic AI Accelerators

NAICS

Semiconductor and Other Electronic Component Manufacturing (3344), Electrical Equipment, Appliance, and Component Manufacturing (335)

SIC

Telephone & Telegraph Apparatus (3661), Electric Lighting & Wiring Equipment (3640)

akta.pro primary industry

Optoelectronics & Photonics (LED/Laser/Optical ICs) (HDAHAAAH)

akta.pro secondary industries

Power Management ICs (PMICs) (HDAHAIAA), Edge AI Hardware & NPUs (SoCs, modules, accelerators) (HDAAAJAC)

Keywords

• Photonic computing
• AI accelerators
• Quantum sensors
• Photonic processors
• Lithium niobate chips

Headquarters

HQ city

Stuttgart

HQ country

Germany

HQ region

Europe

Offices3 records

Markets served

GTM type

B2B

Offering type

Hardware or Manufacturing

Cost components

Technology or R&D, Personnel, Operations, Infrastructure, Supply Chain, Marketing or Sales

Revenue model

1. Hardware Sales — Native Processing Server (NPS): Commercial sales of the Native Processing Server (NPS) — the first commercial photonic processor — to data centers, hyperscalers, HPC centers, and cloud providers. Integrates via standard PCIe interfaces as a co-processor. Production deployments at LRZ and Jülich; first commercial customer IONOS (May 2026).
2. Cloud-based access / Demonstration Licensing: Cloud access to the first-generation NPU launched in September 2024, allowing researchers and developers to experience photonic computing via a web demonstration (handwriting recognition / MNIST inference). Provides a pre-sales evaluation channel.
3. Sensor Product Sales — Q.M 10 Quantum Magnetometer: Sales and pre-orders of the Q.M 10 photonic quantum magnetic field sensor for medical technology applications; 'Early Adopter Program' for researchers and product developers preceding April 2025 market availability.
4. Strategic R&D and Government-Funded Programs: Revenue from government-funded research consortia and programs such as the BMBF-funded LichtBriQ project (TFLN photonic integrated circuits industrialization) and the BMBF/BMFTR-supported pilot line with IMS CHIPS.

Pricing tiers

Go-to-market motion4 records

Distribution channels5 records

Marketing channels10 records

Core offering

Q.ANT develops and manufactures Thin-Film Lithium Niobate (TFLN) photonic processors—the Native Processing Unit (NPU) and Native Processing Server (NPS)—that perform AI and HPC matrix computations natively in the optical domain, delivering up to 30x energy efficiency vs. CMOS. The company also produces photonic quantum sensors, including the Q.M 10 magnetometer for medical technology and biosensing applications. Revenue comes from hardware sales to data centers and HPC centers, plus cloud-access evaluations and government-funded research programs.

Product overview

Q.ANT's offering is a platform-plus-modules architecture organized around its Native Computing platform, which is the company's photonic computing paradigm for fast and sustainable AI and HPC in data centers. The core commercial product is the Native Processing Server (NPS) — the first commercial photonic processor — which houses Q.ANT's Native Processing Units (NPUs), the underlying Thin-Film Lithium Niobate (TFLN) photonic AI accelerators. The NPS ships in two chip generations: the first-generation Native Processing Unit (NPU Gen 1, launched November 2024) tailored for AI inference, and the Q.ANT NPU 2 (Gen 2, unveiled November 2025) delivering up to 50x throughput and 6x lower energy use vs. Gen 1, with support for native nonlinear operations and the foundation for wavelength multiplexing. Complementing the Native Computing business unit is the Native Sensing business unit, which includes the Q.M 10 photonic quantum magnetic field sensor (a sub-product for medical technology and biosensing launched October 2024). The NPS integrates into existing data centers via standard PCIe interfaces as a co-processor alongside CPUs and GPUs. Both products share Q.ANT's proprietary TFLN chip material platform, produced on Q.ANT's own pilot line in Stuttgart via partner IMS CHIPS. The Native Processing Server — Cloud Demonstration at native.qant.com is an add-on developer-access channel to the first-generation NPU.

Differentiator

Problem solved

Functional benefit

Brands

• Native Computing: Business unit focused on photonic processors and AI compute products such as the Native Processing Unit (NPU) and Native Processing Server (NPS).
• Native Sensing

Products and services

• Native Processing Server (NPS) The first commercial photonic processor server, integrating Q.ANT's Native Processing Units (NPUs) for deployment in data centers and HPC environments. Promises up to 30x the energy efficiency of conventional CMOS, integrates via standard PCIe interfaces as a co-processor without specialized cooling, and targets AI training/inference, machine learning, physics simulations, and time-series analysis. Customers include IONOS, LRZ, and Jülich Supercomputing Centre.
• Native Processing Unit (NPU) — Gen 1 The first-generation Native Processing Unit, a photonic AI accelerator built on Q.ANT's proprietary Thin-Film Lithium Niobate (TFLN) material platform. Tailored for AI inference to improve the carbon footprint of next-generation data centers; provides at least 30x energy efficiency vs. CMOS, with one optical element replacing 1,200 transistors for an 8-bit multiplication.
• Q.ANT NPU 2 — Second-Generation Native Processing Unit Second-generation Native Processing Unit unveiled November 2025 on TFLN integrated circuits. Delivers 8 GOPS sustained throughput on nonlinear functions, with one optical element replacing 100–1,000 transistors for the same function. Demonstrated 50x higher throughput in matrix multiplications, 25x faster inference on neural networks, and 6x lower energy consumption vs. Gen 1 at LRZ. Foundation for wavelength multiplexing and next-generation optical scaling.
• Q.M 10 Photonic Quantum Magnetic Field Sensor Photonic quantum magnetic field sensor designed for medical technology and biosensing applications. Uses Q.ANT's Native Sensing approach to perform contactless measurement of fine electrical currents and magnetic fields in the human body, detecting magnetic fields down to 10 Picotesla at room temperature. Enables intuitive prosthetic control through muscle signal pattern recognition with machine learning.

Named customers6 records

Segments5 records

Ideal customer profiles4 records

Technology focussed Yes

API detail

Has API

No

API docs

API detail

Core technology

AI maturity

App detail

Integration3 records

AI capability8 records

Feature6 records

Partnerships

16 partnerships are on record, tiered core, flagship and minor.

• NXAIcoreStrategic or Co-development Partner · 23 June 2026Austrian frontier AI lab commercializing the Extended Long Short-Term Memory (xLSTM) architecture; Q.ANT executed the TiRex time series prediction model on photonic hardware at ISC High Performance 2026. NXAI's commercial TiRex model targets financial market analysis, supply chain optimization, weather forecasting and traffic flow simulation.
• Ludwig Maximilian University of Munich (LMU) — Computer Vision & Learning GroupcoreStrategic or Co-development Partner · 23 June 2026Research collaboration with Professor Dr. Björn Ommer (LMU), the leading researcher behind the development of the stable diffusion model; Q.ANT ran a diffusion model image-to-image synthesis on photonic hardware as a result of this collaboration.
• IONOSflagshipChannel Partner/ Reseller/ Distributor · 19 May 2026Strategic partnership announced May 19, 2026 at re:publica 2026 in Berlin, making IONOS — Europe's leading cloud and hosting provider serving ~6.8 million customers across 17 markets — the first commercial customer for Q.ANT's Native Processing Server (NPS). Marks Q.ANT's transition from research deployments at LRZ and Jülich to commercial photonic AI acceleration in a cloud environment.
• DaisytunercoreTechnology or Integration · 22 April 2026Independent developer partnership — Daisytuner developed a compiler using standard AI toolchains (PyTorch) and successfully compiled/deployed an object detection model directly onto Q.ANT's photonic processor. First AI model from a standard ML framework compiled for photonic hardware.
• Jülich Supercomputing Centre (JSC) / Forschungszentrum JülichflagshipStrategic or Co-development Partner · 17 November 2025Multi-year development partnership launched November 2025 to systematically explore the potential of photonic computing applications at the Jülich Supercomputing Centre.
• Leibniz Supercomputing Centre (LRZ)flagshipStrategic or Co-development Partner · 22 July 2025World's first commercial photonic AI processor deployment in a live production HPC environment (July 2025); running workloads in climate modeling, medical imaging, and fusion energy research. Second-generation NPU evaluation deployed March 2026, demonstrating 50x higher throughput and 6x lower energy consumption.
• IMS CHIPS (Institut für Mikroelektronik Stuttgart)flagshipOEM/ Whitelabel/ Licensing Partner · 24 February 2025Joint operation of Q.ANT's TFLN chip pilot line in Stuttgart, Germany; Q.ANT's photonic chips are produced through this Stuttgart-based partner on refurbished/upcycled CMOS fabrication lines. Established as a blueprint for strengthening chip sovereignty.
• LichtBriQ Consortium (BMBF-funded) — Universität Stuttgart, NTG Neue Technologien, Friedrich-Schiller-Universität Jena, Menlo SystemscoreStrategic or Co-development Partner · 2 January 2025Q.ANT leads the BMBF-funded LichtBriQ consortium (with Universität Stuttgart, Friedrich-Schiller-Universität Jena, NTG Neue Technologien GmbH & Co. KG, and Menlo Systems) to develop an integrated optical construction kit based on TFLN as an industrial platform for photonic integrated circuits (PICs).
• Fraunhofer IPAcoreStrategic or Co-development Partner · 16 October 2024Joint development of an innovative prosthesis sensor module using the Q.M 10 quantum magnetometer; demonstrated at COMPAMED 2024 (Halle 8a/Stand G10) converting emulated muscle signals within milliseconds into precise control commands for a hand prosthesis.
• KPMGcoreImplementation/ SI/ Consulting PartnerM&A advisor for Q.ANT's 68 million euro financing round; structured a management participation scheme compliant with Germany's new Future Financing Act 2024; received the BM&A 'M&A Transaction Advisory' award 2025.
• FestocoreStrategic or Co-development PartnerIndustry partner featured among Q.ANT's 'Strong Alliances' on the homepage logo wall.
• BASFcoreStrategic or Co-development PartnerIndustry partner featured among Q.ANT's 'Strong Alliances' on the homepage logo wall.
• BMBF / BMFTR (Federal Ministry for Research and Technology / Education and Research)coreOthersGerman federal ministry funding Q.ANT's research initiatives including the LichtBriQ project (TFLN photonic integrated circuits industrialization).
• DLR (German Aerospace Center)minorStrategic or Co-development PartnerResearch partner featured among Q.ANT's 'Strong Alliances' on the homepage logo wall.
• phoquantminorStrategic or Co-development PartnerResearch partner featured among Q.ANT's 'Strong Alliances' on the homepage logo wall.
• T-SystemsminorChannel Partner/ Reseller/ DistributorIndustry partner featured among Q.ANT's 'Strong Alliances' on the homepage logo wall; subsidiary of Deutsche Telekom for enterprise IT/digital services.

Scale indicators10 records

Recent moves7 records

Expansion highlights7 records

Broad incumbents

• Xanadu Quantum Technologies: Xanadu is a leading photonic quantum computing company (Borealis) that shares Q.ANT's photonic integrated circuit heritage and adjacent deep-tech positioning, though focused on quantum computing rather than classical photonic AI acceleration. Comparable as European/North American photonics deep-tech players.
• PsiQuantum: PsiQuantum builds a fault-tolerant photonic quantum computer using silicon photonics at scale. While focused on quantum rather than classical compute, PsiQuantum is comparable to Q.ANT as a well-funded photonic deep-tech hardware company addressing compute bottlenecks through optical architectures.
• NVIDIA: NVIDIA is the dominant AI accelerator incumbent (H100/B200 GPUs) whose CUDA software ecosystem defines the competitive landscape Q.ANT is positioning against as a photonic co-processor alternative. NVIDIA's continued performance gains and software moat set the bar Q.ANT must clear on energy efficiency and developer ergonomics to win hyperscaler adoption.

Direct peers

• Lightmatter: Lightmatter is the most directly comparable photonic AI accelerator company, building photonic processors (Envise) and interconnect (Passage) targeting hyperscaler AI workloads. Both Q.ANT and Lightmatter sell photonic co-processors to data centers, compete on energy efficiency vs CMOS, and pursue PCIe integration paths for enterprise adoption.
• Lightelligence: Lightelligence develops optical computing processors (PACE) for AI inference and HPC, using photonic integrated circuits to accelerate matrix multiplication — directly comparable to Q.ANT's NPS/NPU architecture. Both target energy-efficient AI acceleration and compete for hyperscaler and HPC center customers globally.
• Celestial AI: Celestial AI builds the Photonic Fabric platform using silicon photonics for memory interconnect and compute acceleration in AI data centers. It overlaps with Q.ANT on photonic compute primitives for AI workloads and targets similar hyperscaler customers with energy-efficiency value propositions.
• Luminous Computing: Luminous Computing is an early-stage startup building a photonic AI supercomputer using silicon photonics for matrix multiplication, directly comparable to Q.ANT's TFLN-based photonic processors. Both pursue photonic-native AI compute targeting large-scale neural network workloads.

Others

• TRUMPF: TRUMPF is Q.ANT's founding entity and a Series A investor, with deep photonics and laser heritage. As a strategic shareholder and supplier of photonic technology know-how, TRUMPF is an ecosystem participant and enabler rather than a direct competitor in photonic AI compute.
• Fraunhofer Institute (Photonic Technologies / IMS CHIPS): Fraunhofer is Q.ANT's R&D partner for the Q.M 10 prosthesis sensor module and a broader photonic ecosystems ally. As Europe's largest applied research organization with deep photonic chip fabrication capabilities, Fraunhofer is thematically adjacent and a critical ecosystem enabler rather than direct competitor.

Emerging players

• Ayar Labs: Ayar Labs develops optical I/O and in-package photonic interconnect for AI accelerators, using silicon photonics to displace electrical I/O in compute systems. While more focused on interconnect than compute, both Ayar Labs and Q.ANT leverage photonic integrated circuits for AI infrastructure energy efficiency.

Market position

Strengths5 records

Weaknesses4 records

Competitive moat6 records

Key risks5 records

Key highlights7 records

Customer concentration

• YouTubeyoutube.com/channel/UCu4d0Yjo5TtzkJ755zSz61w
• LinkedInlinkedin.com/company/qant
• Xx.com/qantgmbh
• Facebookfacebook.com/qant.gmbh

Compliance2 records

Revenue estimate

Valuation estimate

Number of profiles

Profiles6 records

Funding overview

Funding rounds1 record

Investors11 records

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

M&A

Investments

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Frequently asked questions about Q.ANT