AMD Instinct MI430X OAM
Overview
The AMD Instinct MI430X is the HPC and sovereign-AI member of the CDNA 5 based Instinct MI400 series. It is a distinct part from the AI-focused MI455X rather than a configuration of it: AMD tailors each MI400 part to a precision envelope, and the MI430X keeps full hardware FP64 where the MI455X trades it away for 4-bit and 8-bit throughput. AMD rates it at 288 TFLOPS of hardware FP64, which it puts at about 8.7 times the hardware FP64 of NVIDIA Vera Rubin, and at 9.2 PFLOPS of FP4 for AI work, under a quarter of the MI455X. It carries the same 432 GB of HBM4 at 23.3 TB/s and 256 Work Group Processors with Wave32 execution, and supports UALink for scale-up alongside AMD Infinity Fabric. AMD has not published FP32, FP16, BF16, FP8 or INT8 rates, a TDP or a process node, so those fields stay empty. Announced deployments are the US DOE/ORNL Discovery system, the Alice Recoque exascale system with GENCI and CEA in France, and the Herder system at HLRS. Availability is expected in 2027.
Performance
Peak theoretical throughput by precision type
| Precision | Peak |
|---|---|
FP64 64-bit floating point | 288TFLOPS |
FP32 | No verified data available |
TF32 | No verified data available |
BF16 | No verified data available |
FP16 | No verified data available |
FP8 | No verified data available |
FP6 | No verified data available |
FP4 4-bit floating point | 9,200TFLOPS |
INT8 | No verified data available |
Specifications
Architecture
CDNA 5
Form Factor
OAM
Launch Year
2027
Process Node
No verified data available
Memory
432 GB HBM4
Bandwidth
23,300 GB/s
TDP
No verified data available
Max power
No verified data available
Interconnect
Infinity Fabric
Work Group Processor
256
Spec Confidence
Vendor claimed
Full Specifications
| Compute Engine | |
|---|---|
| Work Group Processor | 256 |
| Compute Units | 256 |
| Memory | |
| Memory | 432 GB |
| Memory Type | HBM4 |
| Bandwidth | 23.3 TB/s |
| Interface Width | No verified data available |
| Interconnect & I/O | |
| GPU-to-GPU | Infinity Fabric |
| Power & Thermal | |
| TDP | No verified data available |
| Max power | No verified data available |
| Enterprise Features | |
| Compute APIs | ROCm, OpenCL, HIP |
| General | |
| Form Factor | OAM |
| Architecture | CDNA 5 |
| Process Node | No verified data available |
| Launch Year | 2027 |
Datasheet & Resources
Flopper Datasheet
AMD Instinct MI430X OAM specifications, generated from our database. Printable.
Vendor product page
AMD documentation for the Instinct MI430X OAM
AMD Instinct MI400 Series GPUs product page
AMD · Latest version
AMD Instinct Documentation
ROCm guides, architecture details
Data Provenance
Every figure traced to a source
Primary Source
- Publisher
- AMD
- Published
- No verified data available
Data Quality
- Spec confidence
- Vendor claimed
- Clock basis
- Boost
- Core precisions with figures
- 2 of 9
- Normalization
- All values in TFLOPS
Compare cloud providers offering on-demand GPU instances for AI training, inference, and HPC workloads.
Browse GPU Cloud ProvidersSimilar GPUs
Frequently Asked Questions
How many TFLOPS does the AMD Instinct MI430X have?
The verified peak figures Flopper holds for the AMD Instinct MI430X are 288 TFLOPS FP64 and 9,200 TFLOPS FP4. Flopper does not currently have verified FP32, FP16 and FP8 throughput figures for it.
What is the power consumption of the AMD Instinct MI430X?
Flopper does not currently have a verified TDP (Thermal Design Power) figure for the AMD Instinct MI430X.
How much memory does the AMD Instinct MI430X have?
The AMD Instinct MI430X is equipped with 432 GB of memory with 23,300 GB/s of memory bandwidth.
What architecture is the AMD Instinct MI430X based on?
The AMD Instinct MI430X is based on the CDNA 5 architecture, launched in 2027.
How much does the AMD Instinct MI430X cost to rent?
Flopper is not tracking any live rental listings for the AMD Instinct MI430X. The provider directory lists the clouds that rent comparable accelerators on demand.
When did the AMD Instinct MI430X come out?
AMD announced the AMD Instinct MI430X for 2027 on the CDNA 5 architecture. It is not yet shipping.
Stay Updated on GPU Releases
Get notified when new GPUs are added or specifications are updated.
No spam, unsubscribe anytime.