Startup Funding: Q3 2024

Numerous new companies burst on the scene in the third quarter of 2024, including startups with plans for customizable RISC-V-based IP for applications from microcontrollers to data centers, high-speed data center interconnects, compute-in-memory LLM inference chips, and surveillance camera SoCs. Although it did not report funding, AheadComputing also launched last quarter to develop RISC-V core IP.

One of the largest rounds of the quarter went to a company developing inspection and metrology equipment. Other areas that drew big spending included SiC wafer processing, chiplet interconnects, and quantum computing. This report highlights 75 companies that collectively raised $2 billion in the third quarter of 2024.

• Summary table
• Chips
• AI hardware
• Test, measurement, inspection
• Materials
• Memory & storage
• Analog & mixed signal
• Photonics & optics
• Quantum computing
• Other technologies

Chips

Akeana emerged from stealth with $100.0M in funding from Kleiner Perkins, Mayfield, Fidelity Ventures, and others. Akeana offers customizable RISC-V-based IP. Its portfolio includes 32-bit RISC-V cores for applications from embedded microcontrollers to edge gateways; 64-bit RISC-V cores and an MMU to support rich operating systems; 64-bit RISC-V cores with higher single-thread performance optimized for next-gen devices, laptops, data centers, and cloud infrastructure; and a computation engine designed to offload Matrix Multiply operations for AI acceleration. Akeana also offers a collection of IP blocks needed to create processor SoCs, including a Coherent Cluster Cache, I/O MMU, and Interrupt Controller IPs, alongside Scalable Mesh and Coherence Hub IP. Founded in 2021, it is based in Santa Clara, California, USA.

DreamBig Semiconductor raised $75.0M in a Series B round led by Samsung Catalyst Fund and the Sutardja Family, joined by new investors Hanwha Corporation, Event Horizon, and Raptor Capital, and existing investors UMC Capital, BRV, Ignite Innovation Fund, Grandfull Fund, and others. DreamBig provides an interconnect fabric platform for chiplet-based design that removes the need for silicon interposers, with open standard interfaces and architecture agnostic support for CPU, AI, accelerator, IO, networking, and memory chiplets that customers can compose in a package. Its architecture weaves HBM interfaces into the chiplet hub fabric, enabling direct access from all chiplets to 3D stacked HBM, DDR, CXL, and SSD memory tiers. DreamBig also offers DPU/SmartNIC networking accelerator chiplets. The startup has optimized its platform for the panel-level packaging technology of Silicon Box. Founded in 2019, it is based in San Jose, California, USA.

Fabric Cryptography raised $33.0M in Series A funding led by Blockchain Capital and 1kx Management, with participation from Offchain Labs, Polygon Labs, and Matter Labs. Fabric Cryptography is developing a Verifiable Processing Unit (VPU), a custom silicon chip that uses an instruction set architecture specific to cryptography that enables it to accelerate any cryptographic algorithm. By using hardware-software co-design techniques, the startup aims to improve the speed and cost of running advanced cryptographic workloads such as zero-knowledge proofs (ZKP), fully homomorphic encryption (FHE), and multi-party computation (MPC). Founded in 2023, it is based in San Francisco, California, USA.

AttoTude launched with the announcement of a $29.0M Series A round with investors including Sutter Hill Ventures, Canaan Partners, and Wing Venture Capital. AttoTude is developing interconnect technology for AI and hyperscale data center applications that it says allows for high-speed data communication at individual lane speeds from 112 Gb/s to 448 Gb/s and beyond. The solution supports high radix, high density architectures and can use existing manufacturing processes. Founded in 2024, it is based in Menlo Park, California, USA.

BigEndian Semiconductors drew $3.0M in seed funding from Vertex Ventures SEA & India and others. BigEndian is developing surveillance camera SoCs for enterprise and consumer applications, with a focus on manageability and a flexible design approach. It anticipates announcing its first chipset soon. Funds will be used for hiring, R&D, and scaling operations. Founded in 2024, it is based in Bangalore, India.

AI hardware

Groq received $640.0M in a Series D round led by BlackRock Private Equity Partners with participation from investors including Neuberger Berman, Type One Ventures, Cisco Investments, Global Brain, and Samsung Catalyst Fund. Groq provides a generative AI inference platform available via the cloud or on-prem that uses its self-developed LPU (language processing units) chips. Founded in 2016, it is based in Mountain View, California, USA.

Fractile emerged from stealth with $15.0M in seed funding led by Kindred Capital, NATO Innovation Fund, and Oxford Science Enterprises, with participation from Inovia Capital, Cocoa.VC, and angel investors. Fractile develops large language model inference chips that combine memory and processing into a single component. By eliminating the time spent moving parameters from memory to a processor, the startup claims it can greatly reduce the amount of time needed to generate a word. Funds will be used for further verification and production of test chips. Founded in 2022, it is based in London, UK.

Rebellions added $15.0M to its Series B round from Saudi Aramco’s Wa’ed Ventures. Rebellions develops domain-specific AI processors along with optimized software. It says it re-architects AI processors to incorporate sophisticated deep learning features through silicon-dedicated DL kernels. It currently offers a product that targets the financial industry and aims to improve trading speeds and reduce latencies for high-frequency trading. It has also developed a chip focused on accelerating AI inference in the data center. It is currently working on its third AI chip, which is focused on large language models for generative AI. Rebellions recently announced plans to merge with Sapeon, another well-funded Korean AI chip startup. Founded in 2020, it is based in Seongnam, South Korea, and has raised over $225M to date.

CogniFiber raised $5.0M in funding led by Chartered Group and Eastern Epic Capital. CogniFiber develops fully-optical analog neuromorphic processors AI that use in-fiber computing. The technology turns a multicore fiber into a programable photonic processor. While data is streaming through the fiber, it also scans all the elements of the physically embedded neural network, enabling compute to be performed while photons are traveling through the fiber. The startup claims its approach provides a massive boost for AI inference while reducing cost, energy, and space. Founded in 2018, it is based in Rosh Haayin, Israel.

Vaire Computing raised $4.5M in seed funding led by led by 7percent Ventures, joined by Seedcamp, Tagus Capital, Page One Ventures, B Heroes, Italian Angels for Growth, Übermorgen Ventures, Yes VC, and numerous individual investors. Vaire is developing chips that implement adiabatic reversible computing. With time-reversible computations, the original inputs can be determined from the output. Since the inputs are not lost when an output is created, the energy that went into the input can be recovered within the system. The startup says this approach results in a design that produces very little heat and thus wastes almost zero power. Vaire is currently targeting applications in generative AI and always-on edge devices and it expects to produce a prototype chip within a year. Founded in 2021, it is based in Bellevue, Washington, USA, and London, UK.

Test, measurement & inspection

Nearfield Instruments raised €135.0M (~$147.4M) in a Series C round led by Walden Catalyst Ventures and Temasek, joined by M&G Investments, Innovation Industries, Invest-NL, and ING Corporate Investments. Nearfield Instruments supplies inspection and metrology equipment, including a semiconductor scanning probe microscopy (SPM) system that can measure 3D surfaces. The company says its Feedforward Trajectory Planner imaging mode can be used for non-destructive in-line process monitoring of 3nm and below gate-all-around FETs (GAA FETs), while a parallel architecture of miniature SPMs increases throughput. The company also offers an in-line, non-destructive subsurface metrology system that combines an acoustic microscopy technique with atomic force microscopy (AFM) technology to ‘listen’ to the sound waves coming through the wafer layers and take nanometer-level measurements of buried features and defects, such as voids, in advanced memory and logic devices. Funds will be used to ramp up production capacity and expand its product portfolio. Founded in 2016 as a spin off from research organization TNO, it is based in Rotterdam, the Netherlands.

Cybord drew $8.7M in Series A funding led by Capri Ventures, with participation from Ocean Azul Partners, IL Ventures,  NextLeap Ventures, and others. Cybord offers a visual AI component analytics platform that inspects, qualifies, and tracks each electronic component in real time to prevent defective, damaged, and counterfeit components from being assembled onto PCBA. Founded in 2018, it is based in Tel Aviv, Israel.

SirenOpt drew $6.6M in seed funding led by Voyager Ventures and Visionaries Club, with participation from Union Labs, Berkeley Skydeck Fund, Wireframe Ventures, Access Industries, Climate Club, and Climate Capital. It also received a $0.3M grant from the National Science Foundation. SirenOpt offers an advanced materials characterization platform that combines cold atmospheric plasma, machine learning, and predictive analytics. The system non-destructively measures and classifies multiple material properties and performance metrics, such as the thickness, density, conductivity, and chemical composition of advanced coatings, thin films, and nanoscale materials used in battery, aerospace, semiconductor, electronics, and other advanced manufacturing industries. Founded in 2022 as a spin out from UC Berkeley, it is based in Oakland, California, USA.

Materials

6K drew $82.0M in the first close of its Series E round from Anzu Partners, Energy Impact Partners, LaunchCapital, Material Impact, and Volta Energy Technologies. 6K uses microwave plasma technology to create alumina, silica, and ceria based powders and specialty particle coatings for chemical mechanical planarization (CMP) polishing of semiconductor wafers. It also applies its technology to the production of additive manufacturing metal powders and lithium-ion battery cathode active materials. Founded in 2014, it is based in North Andover, Massachusetts, USA.

Halo Industries raised $80.0M in a Series B round led by the U.S. Innovative Technology Fund with participation from 8VC and SAIC. Halo Industries processes silicon carbide (SiC) wafers using laser-based tools for grinding, chemical mechanical polishing, and metrology/wafer mapping. The startup claims its processing technologies increase yield and quality while minimizing waste and production cost. It is also developing laser-based wafering and processing technologies for silicon, diamond, sapphire, lithium tantalate, gallium nitride, and other materials. Founded in 2014 as a spin out from Stanford University, it is based in Santa Clara, California, USA.

SweGaN raised nearly €12.0M (~$13.4M) in venture and strategic funding from investors including Navigare Ventures, Wafer Works, RFHIC, Ignite Innovation, BRV Capital Management, and Lifelike Capital. SweGaN manufactures custom gallium nitride on silicon carbide (GaN-on-SiC) epitaxial wafers for RF and power devices used in telecom, defense, and satellite communications. Its approach utilizes a ‘transmorphic heteroepitaxy’ process where vacancy ordering is involved to perfectly transform the in-plane lattice atomic configuration from the SiC substrate to the AlN nucleation layer. The company says its QuanFINE structure combines the strengths of GaN and SiC without the conventional thick C-/Fe- doped buffer layer, bringing the hot channel closer to the heat-sink SiC substrate, reducing the buffer-related traps, and using the ultra wide-bandgap AlN nucleation layer as an effective back-side barrier. A spin off from Linköping University founded in 2014, it is based in Linköping, Sweden.

Memory & storage

RAAAM Memory Technologies was awarded €5.3M (~$5.7M) in grant and equity funding from the European Innovation Council Accelerator. RAAAM has developed an on-chip memory technology it calls Gain-Cell Random Access Memory (GCRAM) that only requires 3 transistors to store a bit of data. The GCRAM bit-cell utilizes decoupled write and read ports, providing native two-ported operation. Proposed as a drop-in SRAM replacement with lower area and power consumption, it is manufacturable using a standard CMOS process. Founded in 2021, it is based in Petach Tikva, Israel.

Analog & mixed signal

Gwanak Analog received KRW 18.0B (~$15.0M) in Series B funding from Samho Green Investment, Korea Development Bank, JB Investment, Mirae Asset Venture, GU Equity, and Seoul Techno Holdings. Gwanak Analog develops analog and power SoCs and ICs for the industrial, consumer, and automotive markets. Its products include an edge AI signal processing chip, power line communication (PLC) modem SoC, display PMIC, multi-sensor signal conditioner, ultrasonic sensor IC, and sensor signal conditioning readout IC with a 24-bit delta-sigma ADC. Founded in 2018, it is based in Seoul, South Korea.

Millibeam received AUD $3.0M (~$2.0M) in financing from Breakthrough Victoria. Millibeam designs mmWave radio chipsets for 5G/6G communications networks. Its chips employ smart frequency switching capabilities that enable it to use both existing 5G spectrum at lower frequencies and the 5G spectrum at millimeter-wave frequencies while improving both energy efficiency and radio link range of terrestrial and non-terrestrial networks. Founded in 2021, it is based in Sydney, Australia.

Photonics & optics

HyperLight raised $37.0M in Series B financing led by Summit Partners, with participation from Xora Innovation and Foothill Ventures. HyperLight provides thin film lithium niobate (TFLN) photonic ICs. Its products include transmitter engine PICs for data center and telecom optical communications, 110 GHz electro-optic modulators, and application-specific PIC solutions through MPW and dedicated wafer runs. Funding will be used to accelerate product development. Founded in 2018, it is based in Cambridge, Massachusetts, USA.

LightSolver received €12.5M (~$13.6M) in grant and equity funding from the European Innovation Council Accelerator. LightSolver has developed a laser-based processing unit for compute-intensive workloads such as CAE, bio-science computations, and optimization problems. The LPU utilizes all-optical coupled lasers that require no electronics to compute, which the startup claims enables a small footprint, scalability, low power requirements, and room temperature operation. Founded in 2020, it is based in Tel Aviv, Israel.

Ephos drew $8.5M in seed funding led by Starlight Ventures, with participation from Collaborative Fund, Exor Ventures, 2100 Ventures, Unruly Capital, Green Sands Equity, Silicon Roundabout Ventures, Club degli Investitori, European Innovation Council, NATO DIANA, and individual investors. Ephos designs and manufactures glass-based photonic chips for quantum computing, data centers, communications, and sensing devices.  It uses a femtosecond laser writing manufacturing process to inscribe waveguides in glass that minimizes signal loss and enables creation of unique 3D device designs. Founded in 2022, it is based in Milan, Italy.

Lightium received $7.0M in seed funding led by Vsquared Ventures and Lakestar. Lightium has developed a thin-film lithium niobate (TFLN) process for manufacturing and packaging of photonic ICs. The startup, which intends to offer foundry services, says its TFLN platform has capabilities to support data transmission speeds exceeding 1.6 Tb/s. It initially targets designs for the telecommunication and data center sectors, but says the platform has applications in markets such as satellite communication, quantum computing, and lidar. Funds will be used to optimize its PDK and accelerate commercialization of its foundry services, which it intends to launch in the beginning of 2025. Founded in 2023, it is based in Zürich, Switzerland.

Quantum computing

Riverlane raised $75.0M in Series C funding led by Planet First Partners, with participation from ETF Partners, EDBI, and existing investors Cambridge Innovation Capital, Amadeus Capital Partners, the UK’s National Security Strategic Investment Fund, and Altair. Riverlane develops a quantum error correction hardware and software stack capable of turning unstable physical qubits into error-free logical qubits. Available for multiple qubit types, it combines an orchestration layer that coordinates and synchronizes complex activities in real time, a decoder that identifies the errors in data and sends corrective instructions, and an interface that processes the qubit readout data from any quantum computer’s control system. Founded in 2016, it is based in Cambridge, UK.

Quantum Circuits Inc. drew $60.0M in Series B financing led by ARCH Venture Partners, F-Prime Capital, Sequoia Capital, and Hither Creek Ventures, joined by Canaan Partners, Fitz Gate Ventures, In-Q-Tel, Osage University Partners, Connecticut Innovations, Tao Capital Partners, and Tribeca Venture Partners. Quantum Circuits is developing full-stack superconducting quantum computers that use a dual-rail qubit with built-in error correction and a modular, scalable architecture. A spin out from Yale University founded in 2015, it is based in New Haven, Connecticut, USA.

Planqc received €50.0M (~$54.2M) in Series A funding led by CATRON Holding and the DeepTech & Climate Fonds, joined by Bayern Kapital, the Max Planck Foundation, UVC Partners, Speedinvest, and private investors, along with a non-dilutive grant from Germany’s Federal Ministry of Education and Research. Planqc is developing a scalable quantum computer based on neutral strontium atoms trapped in optical lattices that can operate at room temperature. Quantum information is then processed with quantum gates based on precisely controlled laser pulses. The startup says its approach can scale to tens of thousands of qubits. Funds will be used to establish a quantum computing cloud service and to develop quantum software for applications in industries such as chemistry, healthcare, climate-tech, automotive, and finance. Based on research from the Max Planck Institute of Quantum Optics and Ludwig Maximilian University of Munich, it was founded in April 2022 and is based in Garching, Germany.

Quantum Source raised a $50.0M Series A round led by Eclipse, with participation from Standard Investments, Level VC, Canon Equity, as well as existing investors Pitango First, Grove Ventures, 10D VC, and Dell Technologies Capital. Quantum Source is developing large-scale, fault-tolerant, server rack-sized photonic quantum computers. The approach uses cavity quantum electrodynamics (cavity-QED) to harness single atoms trapped on a photonic chip to both generate single photons and to implement atom-photon entangling quantum gates, facilitating the construction of complex 3D cluster states that are the backbone of error-correction codes. The startup says that the deterministic nature of the gates minimizes the need for costly and complex feed-forward and switching operations. Founded in 2021, it is based in Nes-Ziona, Israel.

Atom Computing received a $10.0M investment from PensionDenmark, following a DKK 70.0M (~$10.2M) investment last quarter from the Export and Investment Fund of Denmark (EIFO). Atom Computing is building highly scalable, gate-based quantum computers using nuclear spin qubits formed from arrays of optically-trapped neutral atoms. All of the control functions of neutral atom qubits are mediated by light propagating through free space rather than individual electrical cables attached to each qubit, reducing complexity. Additionally, the closed outer electron shell of the alkaline-earth metal atoms used provides insensitivity to environmental perturbations, enabling qubits with >40 second coherence times. The startup has a 100-qubit prototype platform used for quantum algorithm development and is working on building a system with over 1,000 qubits. Funds will support establishment of a European headquarters in Copenhagen. Founded in 2018, it is based in Berkeley, California, USA.

Other tech

LiquidStack drew $20.0M in a Series B extension from Tiger Global. LiquidStack provides liquid cooling for data centers. Its products include direct-to-chip cooling, single and two phase immersion cooling, and modular systems for micro data centers. Founded in 2012, it is based in Carrollton, Texas, USA.

Table

Funds & investors

The Duality Quantum Accelerator accepted five startups focused on quantum science and technology to its latest program.

The ChipStart UK incubator program selected eleven semiconductor startups for its second cohort.