NVIDIA GTC 2025: GB300 AI GPU with 1.4kW power, new details on Rubin AI GPU, CPO tech, and more

NVIDIA is hosting its GPU Technology Conference (GTC) event next week, where we'll be introduced to its beefed-up GB300 "Blackwell Ultra" AI GPUs, new details on its next-gen Rubin R100 AI GPU which will feature next-gen HBM4 memory, new CPO technology, and more.

JP Morgan's outlook for NVIDIA GTC 2025 has Blackwell Ultra GB300 as the highlight of the show, with the new GB300 AI chip featuring a logic structure similar to B200 but with huge capacity increases on HBM and massive increases in power consumption.

• Read more: NVIDIA GB300 'Blackwell Ultra' AI servers: fully liquid-cooled AI clusters
• Read more: NVIDIA CEO confirms Blackwell Ultra, Vera Rubin AI GPUs are on-track
• Read more: NVIDIA's next-gen 'market-grabbing weapon': GB300 AI GPUs to be unveiled at GTC 2025

NVIDIA's new GB300 Blackwell Ultra AI GPU will have up to 288GB of 12-Hi HBM3E memory, and a thermal design power (TDP) that has been increased up to 1.4kW. We are to expect a major performance boost from GB300 of up to 50% faster than GB200 in FP4 computing performance, and JP Morgan expects GB300 to start shipping in Q3 2025.

Blackwell Ultra: A Performance Beast on the Horizon

• HBM Capacity Surge: 288GB, utilizing HBM3e 12-high stacking technology.
• Increased Power Consumption: Thermal Design Power (TDP) reaching 1.4kW.
• Major Performance Boost: Expected to be 50% faster than the B200 in FP4 computing performance.
• Shipment Timeline: Estimated to start shipping in Q3 2025.

NVIDIA has B200 and GB300 on the market right now, with B300 and GB300 to be detailed at GTC 2025 next week, but we'll also be receive more details on the company's next-generation Rubin AI arrchitecture.

NVIDIA's next-gen Rubin AI GPU is expected to feature a dual logic chip structure, where similiar to Blackwell, Rubin will have two chips fabbed on TSMC's new N3 process node. Rubin will also make the huge leap to next-gen HBM4 memory with 8-Hi stack chips with a total capacity of up to 384GB (up from 288GB HBM3E on B300).

Rubin also chows down on even more power than Blackwell, with NVIDIA's next-gen R100 AI GPU expected to have an increased TDP of up to 1.8kW. NVIDIA is expected to upgrade the Vera Arm CPU to TSMC's new N3 process node, potentially using a 2.5D packaging structurer. NVIDIA's new Rubin AI GPU architecture is also expected to move up to a 1.6T network using two ConnectX-9 NICs.

Rubin Platform: The New AI Power Engine for 2026

• Dual Logic Chip Structure: Similar to Blackwell, but equipped with two TSMC N3 process chips.
• HBM4 Memory: Eight HBM4 stacks with a total capacity of 384GB, a 33% increase over Blackwell Ultra.
• Further Power Increase: Estimated TDP of around 1.8kW.
• Vera ARM CPU Upgrade: Expected to migrate to the N3 process, potentially using a 2.5D packaging structure.
• 1.6T Network: Rubin platform may adopt 1.6T networking, featuring two ConnectX-9 NICs.
• NVL144/NVL288 Configurations: Possible introduction of NVL144 and NVL288 rack structures to increase GPU density.
• Production Timeline: There is a possibility of an earlier production timeline, with initial production starting as early as late 2025 or early 2026, and mass shipments expected by Q2 2026.