Weebit Nano (ASX: WBT) has surged more than 300%. What must happen before ReRAM royalties arrive?

Weebit Nano Limited (ASX: WBT) has entered a different stage of its commercial journey after two product customers taped out chips incorporating its resistive random-access memory technology, with one prototype already manufactured and functioning as expected. The milestone starts a testing and qualification process that could take between 12 and 18 months before the products are ready for mass production. WBT shares traded around A$7.50 on July 13, leaving the stock approximately 19% higher over one month and more than 300% above its level a year earlier. That extraordinary re-rating means investors are no longer valuing Weebit simply as a promising memory developer. They are increasingly pricing in the possibility that customer tape-outs will become manufactured products generating recurring royalties.

The challenge is that semiconductor commercialisation moves at a slower pace than share-market expectations. Tape-out is significant because a completed design has been sent for fabrication, but it is not the end of the process. The manufactured silicon must function consistently, survive qualification and enter a customer product that achieves commercial sales.

Why do customer tape-outs represent a different milestone from Weebit’s foundry agreements?

A foundry agreement makes Weebit’s ReRAM technology available within a manufacturing process. It allows chip designers to consider the memory for future system-on-chip products, but it does not necessarily mean a customer has selected the technology for a specific commercial design.

A customer tape-out goes further. The product company has completed a chip design incorporating Weebit ReRAM and released it to a semiconductor manufacturer. At that point, the customer has committed engineering resources, design time and fabrication expenditure to the technology.

Two Weebit customers have now reached that stage. Overlord Labs taped out a chip for a next-generation smart battery-management system at DB HiTek. Another customer has already received first silicon, with initial testing showing that the product and its embedded ReRAM are functioning as expected.

This moves the discussion from whether foundries will offer Weebit ReRAM to whether actual products can progress through qualification. The commercial significance is greater, but the risk has not disappeared. A functional prototype can still encounter performance, reliability, manufacturing-yield or customer-market problems before production begins.

What happens between a successful chip tape-out and the first meaningful ReRAM royalties?

After tape-out, the foundry converts the completed design into physical silicon. Initial chips are then returned to the product customer for functional testing, electrical characterisation and evaluation across expected operating conditions.

Testing must confirm that the ReRAM stores and retrieves data reliably, performs within power and speed specifications and remains stable across temperature, voltage and repeated use. Product companies may also need to test the entire chip rather than the memory module in isolation.

Weebit expects the characterisation and qualification process for its taped-out customer products to take approximately 12 to 18 months. Once qualification is completed, the customer can decide whether to move into mass production.

The timing of royalty revenue will depend on the commercial agreement and the customer’s production schedule. Weebit can receive design-licence, engineering and milestone payments before mass production, but volume royalties generally require finished chips to be manufactured and sold.

This means the first functioning prototype is best understood as the start of a commercial countdown. It reduces technology risk, but it does not establish the eventual production volume. The customer’s product must still address a sufficiently large market and win its own end users.

Why are semiconductor companies looking for alternatives to embedded flash memory?

Electronic devices need non-volatile memory to retain information when power is removed. Embedded flash has performed that role across microcontrollers, automotive chips, industrial devices and consumer electronics for decades.

The problem is that embedded flash becomes increasingly difficult and expensive to manufacture as semiconductor processes move to smaller geometries. It can require additional fabrication steps, high programming voltages and design compromises that become less attractive at advanced nodes.

ReRAM stores information by changing the resistance of a material rather than trapping electrical charge in the way flash memory does. Weebit’s version uses silicon oxide and can be added during the back end of a standard semiconductor-manufacturing process.

That integration model is important because foundries do not want to install specialised equipment or redesign entire fabrication lines for one memory technology. A solution using common materials and a limited number of additional masks may offer a lower-cost route to embedded non-volatile memory.

ReRAM also offers potential advantages in speed, power consumption, endurance and high-temperature operation. These characteristics are relevant to automotive electronics, industrial systems, power-management chips, secure devices and edge-AI processors.

The opportunity does not mean ReRAM will displace every form of flash. Semiconductor companies are conservative because reliability failures can be costly, and competing technologies such as magnetoresistive RAM and phase-change memory are also seeking parts of the emerging-memory market.

How do DB HiTek and SkyWater Technology give Weebit two production-ready routes?

Weebit ReRAM is fully qualified in SkyWater Technology’s 130-nanometre S130 manufacturing process. The qualification allows SkyWater customers to incorporate the memory into new semiconductor products manufactured through the American foundry.

DB HiTek provides another route through its 130-nanometre Bipolar-CMOS-DMOS process. This manufacturing technology is commonly used for mixed-signal, analog and power-management products, markets where embedded memory can support calibration, security, configuration and system intelligence.

Weebit completed industry-standard qualification at DB HiTek, removing an important technical barrier for customers seeking a commercially available ReRAM option. The Overlord Labs tape-out demonstrates how the foundry relationship can progress into a specific product design.

Having more than one qualified foundry reduces dependence on a single manufacturing partner and allows Weebit to address different customers and geographic supply chains. It also increases the chance that a chip designer can access ReRAM through an existing commercial relationship.

The limitation is that qualified availability does not create automatic demand. SkyWater and DB HiTek must promote the technology to customers, while Weebit must support integration and convince product companies that ReRAM offers a better economic solution than familiar alternatives.

What could the onsemi and Texas Instruments agreements contribute to Weebit’s scale?

onsemi licensed Weebit ReRAM for integration into its Treo analog and mixed-signal semiconductor platform. The collaboration targets a 65-nanometre Bipolar-CMOS-DMOS process suitable for automotive, industrial and data-centre applications.

The first test module incorporating Weebit ReRAM was taped out at onsemi’s 300-millimetre production facility during 2025. Testing and qualification are expected to continue through 2026. Successful qualification could give onsemi another embedded-memory option for future product families.

Texas Instruments represents an additional route into high-volume embedded processing. The semiconductor company licensed Weebit ReRAM for advanced manufacturing nodes, giving the technology exposure to a major global supplier with established customer relationships.

These agreements are strategically important because both companies manufacture and sell their own semiconductor products. Unlike a pure-play foundry, an integrated device manufacturer can choose where to deploy the technology within its internal product roadmap.

The financial outcome still depends on implementation decisions. A licence confirms access to the intellectual property, but royalties require Weebit ReRAM to be designed into products that reach production. Development cycles at major semiconductor companies can extend over several years, particularly when automotive or industrial qualification is involved.

Could Weebit’s ReRAM become an important memory technology for edge AI?

Artificial-intelligence systems move large volumes of data between processors and memory. That movement consumes energy and can create a bottleneck, especially in small edge devices that cannot rely on data-centre power or cooling.

ReRAM may support AI applications in two ways. It can provide efficient embedded non-volatile memory for storing models, configuration data and security information. It may also support compute-in-memory architectures where some processing occurs within the memory array, reducing data movement.

Weebit’s technology was selected for a South Korean government-backed analog compute-in-memory programme focused on ultra-low-power artificial intelligence. This gives the company exposure to a potentially significant long-term application beyond conventional flash replacement.

The company intends to use part of its newly raised capital to accelerate AI-related development and commercial offerings. That investment may expand its addressable market, but it also introduces additional research expenditure and technical risk.

AI gives Weebit a powerful market narrative, although the nearer commercial opportunity remains embedded non-volatile memory in conventional semiconductor products. Investors should distinguish between qualified ReRAM available through foundries and experimental compute-in-memory applications that may take longer to produce revenue.

Why did Weebit Nano raise A$102 million when it already held substantial cash?

Weebit held approximately A$72 million at March 31, before receiving proceeds from its capital raising. It subsequently raised A$87 million through institutional and Israeli placements and another A$15 million through a share purchase plan.

The shares were issued at A$4.05, representing a discount to the pre-raising market price. The financing increased Weebit’s share count but gave the company one of the strongest balance sheets among emerging semiconductor intellectual-property developers on the ASX.

The funds are intended to expand commercialisation, accelerate technology transfer to additional foundries, increase research capacity and bring forward AI-memory programmes. Semiconductor development requires sustained expenditure because each process transfer, wafer run and qualification cycle can take considerable time and money.

The size of the raising also carries an important signal. Weebit is preparing for a larger number of simultaneous customer and foundry programmes rather than managing its pipeline with minimal resources. A stronger balance sheet may reassure prospective partners that the company can support multi-year development commitments.

The risk is that capital can be spent well before royalties arrive. More engineers, test wafers and commercial staff improve execution capacity, but they do not guarantee licensing agreements or production adoption. Investors should monitor whether the A$102 million produces a visible increase in signed partners, customer tape-outs and qualified manufacturing processes.

How is the market pricing Weebit Nano after its extraordinary 12-month rally?

WBT traded around A$7.50 during the July 13 session, compared with A$7.41 five trading days earlier and A$6.31 one month earlier. The stock was therefore approximately 1% higher over five sessions and nearly 19% higher over one month, despite considerable daily volatility.

The 52-week range of approximately A$1.82 to A$8.89 captures the scale of the re-rating. At A$7.50, Weebit was only around 16% below the high but more than four times the lower end of the range.

With approximately 240 million shares outstanding, the current price implies a market capitalisation close to A$1.8 billion. That valuation is substantial beside FY2026 revenue guidance of at least A$12 million and demonstrates how much future production success is already reflected in WBT.

Traditional earnings multiples are not useful because Weebit remains loss-making and is investing ahead of royalty revenue. The market is instead assigning value to the probability that ReRAM becomes a widely adopted embedded-memory technology across multiple foundries and semiconductor manufacturers.

This creates asymmetric reactions to new information. Another major licence, successful customer qualification or a production announcement could support the valuation. A delayed foundry agreement, failed qualification or slower customer progress could produce a sharp correction because current revenue provides limited valuation support.

Why does WBT attract unusually strong retail attention and trading volatility?

Weebit offers direct ASX exposure to semiconductor memory at a time when artificial intelligence, supply-chain security and advanced manufacturing have become dominant investment themes. Few Australian-listed companies provide a similarly focused connection to global chip development.

The company also has a simple long-term proposition that resonates with retail investors: embedded flash is becoming harder to scale, and ReRAM could become one of its successors. Relationships with Texas Instruments, onsemi, DB HiTek and SkyWater Technology add recognisable industry validation.

Sentiment is nevertheless divided. Supporters view customer tape-outs and major semiconductor licences as evidence that Weebit is approaching the royalty stage. More cautious investors focus on the A$1.8 billion valuation, uncertain production timelines and the gap between current revenue and future expectations.

The share price reflects this disagreement through large daily moves. WBT traded through a range of almost A$1 during the July 10 session before closing 7.1% higher. Similar volatility can occur even without a new company announcement because traders respond to semiconductor sentiment, institutional positioning and expectations around the next licence.

Retail investors should avoid confusing share-price momentum with a shorter commercial timetable. The stock can move in minutes, while qualification and production decisions unfold over quarters or years.

What milestones will determine whether Weebit reaches volume production?

The first milestone is the delivery and testing of the Overlord Labs chip manufactured at DB HiTek. Initial functional confirmation would move the product into deeper characterisation and qualification.

The second customer’s functioning prototype must continue through reliability and performance testing. Completion of that process could create Weebit’s first customer product ready for mass production.

Qualification of the ReRAM module within onsemi’s Treo platform is another important event. Success would make the technology available across a broad analog and mixed-signal product environment.

Investors are also waiting for additional foundry or integrated-device-manufacturer licences. A previously targeted third new manufacturing agreement slipped from 2025 into 2026, making execution against that objective an important measure of commercial momentum.

The June-quarter report will show whether Weebit met its FY2026 revenue guidance of at least A$12 million and how much cash it used during the commercial expansion. Revenue may remain uneven because licence and milestone payments do not follow a smooth quarterly schedule.

The milestone that ultimately changes the company’s financial structure is mass production. Once customer products containing Weebit ReRAM are manufactured in meaningful volumes, the business can begin demonstrating the royalty leverage embedded in its licensing model.

Key takeaways from the Weebit Nano production and WBT share-price roadmap

• Two product customers have taped out chips containing Weebit ReRAM, and one customer already has a functioning prototype.
• Customer testing, characterisation and qualification may take 12 to 18 months before the products can move into mass production.
• Overlord Labs is integrating Weebit ReRAM into a smart battery-management chip manufactured through DB HiTek.
• Qualified manufacturing routes at DB HiTek and SkyWater Technology give product companies access to production-ready ReRAM processes.
• Agreements with onsemi and Texas Instruments provide exposure to larger semiconductor product portfolios, but production royalties are not guaranteed.
• Weebit raised approximately A$102 million to expand commercialisation, foundry transfers and AI-memory development.
• WBT has risen more than 300% over one year and carries a market capitalisation close to A$1.8 billion, creating significant sensitivity to qualification delays or new commercial wins.