Next-generation PCIe 6.0 SSDs are officially on the horizon for the enterprise market, with major storage makers targeting launches between 2026 and 2027. The push is driven by explosive growth in AI workloads, where every microsecond of latency and every watt of power matters at scale. Expect big gains in throughput, IOPS, capacity, and efficiency—alongside new form factors and memory technologies engineered for dense, high-performance data centers.
At a recent industry summit in Shenzhen, multiple vendors outlined concrete roadmaps for PCIe Gen6 storage. Here’s what’s coming and when:
– Samsung is preparing a wave of PCIe 6.0 and CXL 3.1 products for 2026. Its PM1763 Gen6 SSD is slated for early 2026 with double the performance over prior-generation solutions, significantly higher energy efficiency, and a 25W power envelope to keep rack-level power budgets in check. The company is also advancing memory-semantic storage with next-gen CMM-D solutions on PCIe 6.0, planned for next year. On the capacity front, Samsung’s 256 TB Gen5 drives are starting to roll out, while an enormous 512 TB PCIe Gen6 SSD in the EDSFF 1T form factor is targeted for around 2027. In parallel, Samsung is prepping its 7th-generation Z-NAND with GIDS, expected to arrive next year, aiming to slash latency for ultra-demanding workloads.
– InnoGrit is focusing squarely on AI-centric enterprise storage with a PCIe 6.0 platform designed to deliver up to 25 million IOPS on 512-byte random reads. The company expects to ship its first PCIe Gen6 SSDs in 2026, targeting the low-latency, high-concurrency profiles common in AI inference, vector databases, and real-time analytics.
– Silicon Motion showcased the SM8466 PCIe Gen6 SSD controller, built to enable up to 28 GB/s throughput and support capacities scaling to 512 TB. Enterprise deployments using controllers like the SM8466 are expected to appear in 2026, laying the foundation for next-gen all-flash arrays and AI-optimized storage nodes.
For IT leaders planning refresh cycles, the message is clear: the 2026–2027 window will be a pivotal inflection point for enterprise storage. PCIe 6.0 effectively doubles the available bandwidth over Gen5, enabling faster model training, quicker checkpointing, and higher data ingestion rates across AI pipelines. With denser EDSFF 1T drives hitting 512 TB per device, organizations can dramatically increase storage per rack unit, simplify fleets, and reduce total cost of ownership by consolidating arrays without sacrificing performance.
Equally important is efficiency. Hitting multi-GB/s per lane while keeping power to roughly 25W per drive helps maintain thermal equilibrium in densely packed servers. That’s crucial as AI clusters scale out—and as budgets and sustainability targets face new scrutiny.
It’s also worth noting what’s not arriving soon: consumer PCIe 6.0 SSDs. Although the enterprise ecosystem will begin deploying Gen6 hardware in 2026, client-grade drives aren’t likely to hit the market until roughly 2028–2029. The early focus is on data centers, where the bandwidth uplift, latency improvements, and massive capacities deliver immediate, measurable ROI.
Bottom line: PCIe 6.0 SSDs are set to transform enterprise storage starting in 2026. Expect double the throughput, huge gains in IOPS, far denser drives up to 512 TB, and smarter power profiles—all tuned for AI-heavy data centers. If you’re mapping out your next storage upgrade, align procurement, firmware readiness, and rack design now to capitalize on the first wave of Gen6 deployments.
