NAND Flash wafer supply tightens as Artificial Intelligence demand lifts prices

TrendForce reports strong November 2025 demand for NAND Flash driven by Artificial Intelligence applications and enterprise SSD orders. Wafer supply tightened and November contract prices rose by monthly averages of 20% to over 60% across product categories.

TrendForce’s latest research indicates that demand for NAND Flash remained strong in November 2025, with Artificial Intelligence applications and solid enterprise SSD orders cited as the primary demand drivers. The report highlights sustained customer interest across segments, which underpinned continued ordering activity throughout the month. November’s demand dynamics reflected a market environment where new workloads and storage procurement kept consumption elevated relative to prior periods.

Suppliers responded by reallocating capacity toward higher-margin enterprise and premium products while accelerating the phase-out of old-node capacity. This strategic shift concentrated manufacturing resources on newer, more profitable product lines and reduced available output for mainstream and legacy wafers. TrendForce describes this capacity rebalancing as an intentional prioritization of premium segments that narrowed the supply available to other density and mainstream categories.

The combined effect of robust demand and supplier capacity focus led to a further tightening of wafer supply in November 2025 and a sharp rise in contract prices for mainstream wafers. TrendForce reports that monthly average price increases varied from 20% to over 60% across all product categories, and that the price surge quickly expanded to all density segments. The research frames the November movement as broad and rapid, affecting multiple product tiers as supply constraints and shifting capacity allocations intersected with stronger order flow.

55

Impact Score

Introducing Mistral 3: open artificial intelligence models

Mistral 3 is a family of open, multimodal and multilingual Artificial Intelligence models that includes three Ministral edge models and a sparse Mistral Large 3 trained with 41B active and 675B total parameters, released under the Apache 2.0 license.

NVIDIA and Mistral Artificial Intelligence partner to accelerate new family of open models

NVIDIA and Mistral Artificial Intelligence announced a partnership to optimize the Mistral 3 family of open-source multilingual, multimodal models across NVIDIA supercomputing and edge platforms. The collaboration highlights Mistral Large 3, a mixture-of-experts model designed to improve efficiency and accuracy for enterprise artificial intelligence deployments starting Tuesday, Dec. 2.

Contact Us

Got questions? Use the form to contact us.

Contact Form

Clicking next sends a verification code to your email. After verifying, you can enter your message.