Q3 2026 Lead Time Report Highlights

The electronic components market entered Q3 2026 facing a now familiar imbalance. As demand for artificial intelligence infrastructure keeps growing, it is outpacing the ability of large portions of the supporting supply chain to adjust. The result is a market increasingly divided between products exposed to AI infrastructure and more traditional commodity components.  

Memory remains at the center of this shift. Suppliers continue prioritizing HBM and server products, restricting capacity available for conventional DRAM and other applications. But Q3’s defining development is that the market is no longer witnessing a simple memory shortage.  

AI infrastructure requires an entire ecosystem around the processor: power conversion, cooling, storage, networking, connectors, capacitors, embedded controllers, and high-performance circuit boards. As deployment scales, pressure is moving downstream into each of those supporting technologies.  

Q3 2026 Lead Time & Price Takeaways

Three themes define Q3’s conclusions. First, AI’s influence now reaches nearly every layer of the supply chain, from raw materials to bleeding-edge chips. Even product families with historically stable availability are seeing longer lead times as manufacturers prioritize high-margin components.  

Second, price increases have become routine rather than the exception. A trend that began with memory hikes by Samsung, SK Hynix, and Micron in late 2025 has since spread across nearly every component sector. Buyers have received 27 price increase notices from 20 major manufacturers, with many of those already on their second increase of the year.  

Finally, costs are rising from the bottom up. Higher prices for copper and other metals, along with rising packaging, logistics, and energy costs are driving up manufacturing expenses even for parts that don’t compete directly with AI for allocation. Tariffs, export controls, and stricter supplier terms add to the pressure, meaning buyers can no longer judge lead time and price risk by component category alone.  

Memory

The memory category continues to see AI and cloud applications drive enterprise demand while consumer demand remains relatively weak due to high prices and limited availability. Increasingly, memory supply is being determined by application and profitability.

Volatile memory remains one of the most constrained electronic component categories. With suppliers prioritizing capacity toward HBM and server DRAM, conventional memory applications have less flexibility. DDR4, LPDDR5, and select DDR5 products remain on allocation almost across the board. Micron’s lead times for several of its technologies have reached 52 weeks and most other manufacturers aren’t far behind.  

Meanwhile, South Korea has reported that DRAM prices have climbed 12.5x in three years, prompting Gartner’s Rajeev Rajput to warn that “memflation” could destroy or delay non-AI demand into 2028. Experts fear Nvidia’s 15% server hike may help push HBM prices upward by another 50% in 2027.  

The story for NAND is much the same. SSD lines remain on allocation from most manufacturers, with lead times ranging from 24 to 52 weeks and trending longer. Thanks to emerging data center demand, enterprise SSDs are expected to surpass 50% of all NAND shipments by the end of 2026.  

Passives

Passive components were among the clearest examples of AI demand spreading downstream during Q3. Per TrendForce, standard MLCC lead times are hovering at roughly 14 to 18 weeks while high-specification products are stretching toward five to 10 months. Some Samsung Electro-Mechanics high-capacitance MLCCs are reportedly approaching 40 weeks.

With a single AI server rack consuming upwards of 440,000 high-end MLCCs and the category having extremely limited expansion capacity to meet this demand, some suppliers have raised prices by 30% or more. Tantalum prices have also surged, further increasing cost pressures.

Meanwhile, high-capacitance and high-voltage MLCCs used in servers and high-performance power supplies are experiencing significantly tighter conditions than standard products.

Power, Analog, and Embedded Processing

Higher power density in AI server racks creates substantially greater requirements for voltage regulation, conversion, protection, and thermal management. This demand is contributing to tighter market conditions across MOSFETs, diodes, rectifiers, wide-bandgap devices, and power management ICs.  

The advanced analog category follows a similar pattern. Power-management ICs, data converters, operational amplifiers, and interfaces are seeing increasing price pressure, while certain suppliers are posting longer lead times. Supplier pricing behavior reinforces that trend with numerous price increases in the last few weeks.

Meanwhile, the MCU market is not experiencing a universal shortage, but Q3 revealed that availability is becoming increasingly specific to manufacturer, architecture, and package. Lead times at STMicroelectronics have extended with multiple product families reaching 52 weeks, demonstrating why procurement teams should look beyond broad product family averages when evaluating embedded processing exposure.

Materials, Interconnect, and More

Beyond the categories already discussed, three areas deserve attention:

• Raw materials: Input costs for copper and other metals are rising with trickle-down ramifications for numerous products. Dysprosium and yttrium are now under export controls while polysilicon is subject to tariffs.  ‍
• Interconnect: Supply is broadly stable, but multiple suppliers are extending lead times. With AI infrastructure demand already expected to keep pricing firm through Q4, input costs deserve a close eye.  ‍
• Electro-mechanical: Thermal management has emerged as a key bottleneck for AI applications as compute density increases cooling requirements. Both lead times and pricing are expected to rise through Q4.  

Is AI demand creating a PCB material shortage?

There is strong evidence that AI is fundamentally changing the PCB market, but the effects must be described carefully. PCB & Design Fab reported in October that industry growth has shifted away from traditional PCs and smartphones toward advanced IC package substrates and high-layer-count multilayer boards used in AI hardware. The publication notes that resource allocation caused by AI has spread beyond advanced AI components into the broader electronics supply chain during the second half of 2026.  

This doesn’t necessarily amount to a blanket shortage yet, but it is clear AI infrastructure is changing which PCB technologies attract capacity and investment. Procurement teams reliant on conventional materials should therefore monitor supplier commitments and keep qualified alternatives ready, even when a material has not formally reached end-of-life status.

Quarter-over-Quarter Comparison

In Q2, memory allocation was the most visible constraint, though pressure was building across power, passives, interconnect, and thermal management. Raw material inflation was also starting to emerge, but most other categories still had reasonably stable availability.  

By Q3, those secondary pressures amounted to measurable supply chain issues and repeated supplier price increases. High-end MLCC lead times stretched toward 10 months, select MCU and FPGA discrete products reached 40–52 weeks, and discrete and power-management products showed broader upward lead-time trends. Thermal and power-conversion products have also tightened as they prove to be a difficult AI bottleneck.  

Looking back to a year ago, DRAM constraints were only beginning to surface. Now, shortages span nearly the entire AI hardware stack. Figures in Q3 portray an increasingly fragmented market with lead times, allocation risk, and pricing power primarily being determined by AI exposure.  

Q4 is unlikely to bring broad relief. TrendForce’s September forecast calls for DRAM contract prices to rise another 10% to 15% quarter over quarter while NAND Flash rises 15% to 20% in the same frame. Enterprise SSD price increases are also expected to accelerate as cloud service providers place more orders.  

For procurement teams, the most significant challenge will be identifying where AI demand intersects with your BOM. As shown in Q3, the greatest exposure may no longer be the obvious GPU or memory component. It could instead be the capacitor, power device, MCU package, connector, cooling component, storage device, or PCB material supporting the system. Teams that accurately map exposure now will be far better positioned than those reacting when price increases or allocation notices arrive.