What changed #
On 28 July 2026, two MLCC leaders announced price increases on the same day. Taiwan media reported that Taiyo Yuden notified customers of a second 2026 price adjustment, effective 1 September, after its first round took effect on 1 May. The notice also stated that, even after the increase, the company cannot fully guarantee original lead times. Separately, Korea media reported that Samsung Electro-Mechanics told customers all MLCC products would rise 30%, applied to shipments from 1 August. Murata, the largest MLCC maker, has already raised prices three times this year.
This is a new, dated supply event with specific effective dates, vendor notices and price or lead-time changes — distinct from the earlier July market snapshot that only noted MLCC broadly tightening.
Why buyers should care #
The driver is structural, not cyclical: AI servers consume far more MLCCs than traditional systems. Reported industry figures put a standard server at roughly 2,200 MLCCs, an NVIDIA NVL72 rack at 440,000–600,000, and the next-generation Rubin platform at 570,000 or more.
High-capacity MLCCs consume four to five times the wafer capacity of a standard part, and ultra-high-capacity yields are reported near 40%, which indirectly squeezes consumer-grade supply. The result is a split market: high-end lines run near full, while mid and low consumer lines sit lower. High-end MLCC lead times have stretched from a historical average of about 8 weeks to more than 20 weeks, with some models at 26–40 weeks. Late-June book-to-bill ratios for the three leaders were reported at 1.25–1.31, all at multi-year highs.
For almost every BOM, MLCC is a high-count, low-unit-cost line item — so a 30% list-price move and uncertain lead times hit both cost and continuity.
Affected component areas #
- Passive components — MLCC across dielectric, case size, voltage and tolerance classes; high-capacity parts first
- AI server, HPC and networking platforms (highest MLCC intensity)
- Consumer electronics, via capacity reallocation away from standard parts
- Automotive and industrial designs using high-reliability MLCC
- Power, RF and decoupling sections where MLCC filtering is pervasive
RFQ and sourcing checks #
| Risk signal | Buyer action |
|---|---|
| List-price move (Samsung +30% from 1 Aug; Taiyo second round from 1 Sep) | Request an updated quotation for target quantity; confirm whether the new price applies to open orders and unshipped POs |
| Lead time not guaranteed (Taiyo); high-end 20+ weeks | Confirm the committed delivery date on current orders; avoid single-source for critical lines |
| Capacity shifted to high-cap parts | Specify dielectric, case size, voltage, tolerance and termination exactly in the RFQ; vague "MLCC" requests may be deprioritized |
| BB ratio >1.25, utilization split | Qualify alternate sources and buffer stock for long-life and safety-related programs |
| Q3 consumer peak season | Front-load orders planned for September–October build to avoid the tightest window |
Practical takeaway #
Treat the two effective dates as action points: confirm by early August how Samsung's 30% move affects your open quotations, and by early September how Taiyo's second round and its "lead time not guaranteed" stance hit your committed deliveries. Write exact specifications into every MLCC RFQ, and for AI-server and long-life programs secure allocation or buffer stock now rather than at build time.
Industry estimates from JPMorgan and Guosen Securities point to a supply gap persisting to 2028, with cumulative ASP gains estimated at 50–60% — use these as a planning horizon, not as a quote. LimChip can confirm live stock, date code and lot traceability for MLCC lines such as the Murata GRM188R71H103KA01D by RFQ.
Need stock, date-code or package confirmation?
Send the part number, quantity, target date code and packaging requirements. LimChip will check available lots and RFQ details before you place the order.
Send RFQ