Radeon Instinct MI455X vs H100 SXM5

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	Hopper (2022−2024)	CDNA 5.0 (2026)
GPU code name	GH100	MI450 256CU
Market segment	Workstation	Workstation
Release date	22 March 2022 (4 years ago)	23 July 2026 (recently)

Detailed specifications

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

Pipelines / CUDA cores	8448	32768
Core clock speed	1065 MHz	1000 MHz
Boost clock speed	1780 MHz	2400 MHz
Number of transistors	80,000 million	320,000 million
Manufacturing process technology	4 nm	2 nm
Power consumption (TDP)	700 Watt	1400 Watt
Texture fill rate	939.8	2,457.6
Floating-point processing power	30.07 TFLOPS	157.3 TFLOPS
ROPs	24	no data
TMUs	528	1024
Tensor Cores	528	no data
L1 Cache	24.8 MB	8 MB
L2 Cache	50 MB	192 MB

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
Width	no data	EAM Module
Supplementary power connectors	8-pin EPS	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	HBM3	HBM4
Maximum RAM amount	80 GB	432 GB
Memory bus width	5120 Bit	24576 Bit
Memory clock speed	1500 MHz	1900 MHz
Memory bandwidth	1,920 GB/s	23.3 TB/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

No outputs

API and SDK support

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

DirectX	N/A	N/A
Shader Model	N/A	N/A
OpenGL	N/A	N/A
OpenCL	3.0	3.0
Vulkan	N/A	N/A
CUDA	9.0	-
DLSS	+	-

Pros & cons summary

---

Recency	22 March 2022	23 July 2026
Maximum RAM amount	80 GB	432 GB
Chip lithography	4 nm	2 nm
Power consumption (TDP)	700 Watt	1400 Watt