Lexar NM1090 PRO PCIe 5.0 NVMe SSD 2TB

The Lexar NM1090 PRO is a high-performance M.2 2280 NVMe SSD built on the PCIe 5.0 x4 interface, leveraging the latest Gen5 host controller and 3D NAND architecture to achieve sequential read speeds of up to 14,000 MB/s and sequential write speeds of up to 12,000 MB/s. These figures represent a near-doubling of throughput compared to PCIe Gen4 flagship drives, making the NM1090 PRO purpose-built for platforms requiring extreme bandwidth such as AMD Ryzen 9000 series, Intel Core Ultra 200 series, and professional workstation chipsets that expose full Gen5 lanes to the M.2 slot. The drive ships in a 2TB capacity point with an onboard DRAM cache that sustains peak performance across large sequential transfers and mixed random I/O workloads.

The NM1090 PRO targets professional content creators, 3D rendering artists, video editors working with 8K RAW footage, machine learning engineers, and data scientists who routinely stage large datasets directly on local NVMe storage. Its random read IOPS exceed 1,500,000 and random write IOPS reach 1,400,000, ensuring that even metadata-intensive workloads such as large project file indexing, asset management databases, and virtual machine boot storms are handled without perceptible latency spikes. An integrated aluminum heat spreader with thermal throttle management keeps junction temperatures within specification under sustained loads without requiring a separate aftermarket cooler.

Launched in September 2025, the NM1090 PRO fills a demanding segment of the professional storage market where PCIe Gen4 drives can no longer satisfy bandwidth-hungry applications. The drive supports NVMe 2.0 protocol features including improved queuing and namespace management, and is compatible with both consumer high-end workstations and select enterprise edge-compute platforms that provision Gen5 M.2 slots. Its combination of raw throughput, low-latency random access, and robust thermal design makes it an immediately practical upgrade for studios, post-production houses, and scientific computing environments running bandwidth-limited pipelines.

Ideal for

• 8K and 12K RAW video editing with DaVinci Resolve or Adobe Premiere Pro, where sustained 12,000+ MB/s write bandwidth eliminates proxy workflows
• Machine learning dataset staging for local GPU training rigs, allowing rapid iteration over large image or NLP corpora stored directly on NVMe
• 3D scene rendering and simulation scratch storage for Blender, Cinema 4D, and Houdini pipelines requiring fast asset streaming and cache I/O
• Virtual machine hosting on high-end workstations, leveraging 1.5M+ random read IOPS to support multiple concurrent VM boot and live migration operations
• Scientific computing and bioinformatics pipelines that stage reference genome databases or seismic datasets locally to avoid network storage bottlenecks
• Game development build systems and asset compilation pipelines where fast shader compilation and large asset pack reads benefit from Gen5 sequential throughput

Technical specifications

Available from Omnixon Global. Submit an RFQ and our team will confirm configuration and availability for your order.