ECC is supported on Linux only, which AMD states in the specification table itself as 'Yes (Linux Only)'. This is the only AMD part outside the Instinct line that publishes both an FP8 rate and a structured-sparsity figure, and the dense left-hand column is what is stored here. AMD publishes no lithography and no memory clock for this card, and no FP64 or BF16 rate for any RDNA 4 part.

AMD Radeon AI PRO R9700 32GB

RDNA 4 PCIe 2025 Spec confidence: Official
FP8 (dense)
383
TFLOPS
FP32
48
TFLOPS
VRAM
32 GB
GDDR6
Bandwidth
640 GB/s
memory
TDP
300 W
159.3 TFLOPS/kW
Download datasheet

Overview

The Radeon AI PRO R9700 is AMD's answer to a professional buyer who wants 32 GB of local memory without paying Instinct prices. It is the full 64 compute unit Navi 48 configuration with 128 AI accelerators, running at up to 2920 MHz for 47.8 TFLOPS of FP32 vector, 191 TFLOPS of dense FP16 matrix and 383 TFLOPS of dense FP8 in both E5M2 and E4M3 encodings. Memory is 32 GB of GDDR6 on a 256-bit bus at 640 GB/s with ECC on Linux, in a 300 W double-slot card fed by a single 12V-2x6 connector. Against the workstation card it replaces, the 48 GB Radeon PRO W7900, it trades a third of the capacity for triple the FP16 matrix throughput and for FP8 support the older card does not have at all.

Performance

Peak theoretical throughput by precision type

PrecisionDense2:4 Sparse
FP64
No verified data available Structured sparsity is a tensor-core feature; this vector precision has no sparse form
FP32
32-bit floating point
48TFLOPS Structured sparsity is a tensor-core feature; this vector precision has no sparse form
TF32
No verified data available The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision
BF16
No verified data available The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision
FP16
16-bit floating point
191TFLOPS 383TFLOPS
FP8
8-bit floating point
383TFLOPS 766TFLOPS
FP6
No verified data available The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision
FP4
No verified data available The part supports 2:4 sparsity, but the vendor publishes no sparse figure for this precision
INT8
8-bit integer
383TOPS 766TOPS
INT4
4-bit integer
766TOPS 1,531TOPS

The Radeon AI PRO R9700 32GB in the GPU landscape

Peak FP16 TFLOPS (dense) against TDP, single-GPU parts tracked by Flopper

06001,2001,8002,4003,0000 W250 W500 W750 W1000 W1250 W1500 WInstinct MI355XRadeon AI PRO R9700 32GB

Higher and further left is better: more half-precision throughput for less power.

Specifications

Architecture

RDNA 4

Form Factor

PCIe

Launch Year

2025

Process Node

No verified data available

Memory

32 GB GDDR6

Bandwidth

640 GB/s

TDP

300 W

Max Power

300 W

Transistors

53.9 billion

Stream Processors

4,096

Spec Confidence

Official

Full Specifications

Compute Engine
Stream Processors 4,096
Tensor Cores 128 (RDNA 4)
Compute Units 64
Boost Clock 2.92 GHz
Chip Design
Transistors 53.9 billion
Memory
VRAM 32 GB
Memory Type GDDR6
Bandwidth 640 GB/s
Interface Width 256-bit
Interconnect & I/O
PCIe Gen 5 x16
Cache
Infinity Cache 64 MB
Power & Thermal
TDP 300 W
Max Board Power 300 W
Power Connector 12V-2x6
Cooling Active
Enterprise Features
ECC Memory Yes
Sparsity Yes
Compute APIs ROCm, OpenCL, HIP
Physical & Media
Width 2-slot
General
Form Factor PCIe
Architecture RDNA 4
Process Node No verified data available
Launch Year 2025

Datasheet & Resources

Data Provenance

Every figure traced to a source

Primary Source

Publisher
AMD
Published
2025-05-20

Data Quality

Spec confidence
Official
Clock basis
Boost
Core precisions with figures
4 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 Providers

Similar GPUs

Frequently Asked Questions

How many TFLOPS does the AMD Radeon AI PRO R9700 have?

The AMD Radeon AI PRO R9700 delivers 48 TFLOPS FP32, 191 TFLOPS FP16 and 383 TFLOPS FP8 at peak.

What is the power consumption of the AMD Radeon AI PRO R9700?

The AMD Radeon AI PRO R9700 has a TDP (Thermal Design Power) rating of 300 watts.

How much memory does the AMD Radeon AI PRO R9700 have?

The AMD Radeon AI PRO R9700 is equipped with 32 GB of VRAM with 640 GB/s of memory bandwidth.

What architecture is the AMD Radeon AI PRO R9700 based on?

The AMD Radeon AI PRO R9700 is based on the RDNA 4 architecture, launched in 2025.

Stay Updated on GPU Releases

Get notified when new GPUs are added or specifications are updated.

Loading verification...

No spam, unsubscribe anytime.

Back to GPUs