GB10 vs Data Center GPU Max Subsystem

Primary details

GPU architecture, market segment, value for money and other general parameters compared.

Place in the ranking	not rated	not rated
Place by popularity	not in top-100	not in top-100
Architecture	Generation 12.5 (2021−2023)	Blackwell (2024−2025)
GPU code name	Ponte Vecchio	GB10
Market segment	Workstation	Workstation
Release date	10 January 2023 (2 years ago)	27 August 2025 (recently)

Detailed specifications

General parameters such as number of shaders, GPU core base clock and boost clock speeds, manufacturing process, texturing and calculation speed. Note that power consumption of some graphics cards can well exceed their nominal TDP, especially when overclocked.

Pipelines / CUDA cores	16384 ×4	6144
Core clock speed	900 MHz	1665 MHz
Boost clock speed	1600 MHz	2525 MHz
Number of transistors	100,000 million	no data
Manufacturing process technology	10 nm	3 nm
Power consumption (TDP)	2400 Watt	no data
Texture fill rate	1,638 ×4	969.6
Floating-point processing power	52.43 TFLOPS ×4	31.03 TFLOPS
ROPs	no data	48
TMUs	1024 ×4	384
Tensor Cores	1024 ×4	384
Ray Tracing Cores	128 ×4	48

Form factor & compatibility

Information on compatibility with other computer components. Useful when choosing a future computer configuration or upgrading an existing one. For desktop graphics cards it's interface and bus (motherboard compatibility), additional power connectors (power supply compatibility).

Interface	PCIe 5.0 x16	PCIe 5.0 x16
Length	267 mm	150 mm
Width	2-slot	IGP
Supplementary power connectors	1x 16-pin	None

VRAM capacity and type

Parameters of VRAM installed: its type, size, bus, clock and resulting bandwidth. Integrated GPUs have no dedicated video RAM and use a shared part of system RAM.

Memory type	HBM2e	LPDDR5X
Maximum RAM amount	128 GB ×4	128 GB
Memory bus width	8192 Bit ×4	256 Bit
Memory clock speed	1565 MHz	1067 MHz
Memory bandwidth	3,205 GB/s ×4	273.2 GB/s
Resizable BAR	-	+

Connectivity and outputs

This section shows the types and number of video connectors on each GPU. The data applies specifically to desktop reference models (for example, NVIDIA’s Founders Edition). OEM partners often modify both the number and types of ports. On notebook GPUs, video‐output options are determined by the laptop’s design rather than the graphics chip itself.

Display Connectors	No outputs	1x HDMI
HDMI	-	+

API and SDK compatibility

List of supported 3D and general-purpose computing APIs, including their specific versions.

DirectX	12 (12_1)	N/A
Shader Model	6.6	N/A
OpenGL	4.6	N/A
OpenCL	3.0	3.0
Vulkan	N/A	N/A
CUDA	-	10.1
DLSS	+	+

Pros & cons summary

---

Recency	10 January 2023	27 August 2025
Chip lithography	10 nm	3 nm

Vote for your favorite

Do you think we are right or mistaken in our choice? Vote by clicking "Like" button near your favorite graphics card.

---