AMD · 2025-03-01

Radeon RX 9070

The AMD Radeon RX 9070 is a powerful graphics card based on the AMD RDNA™ 4 architecture. It offers 56 unified compute units, 16GB of video memory, and a boost clock of up to 2.52 GHz. With support for ray tracing and AI acceleration, it delivers excellent gaming performance and visual quality.

Radeon RX 9070 — illustration of the card's form factor
VRAM
16 GB
FP32 TFLOPS
36.1 TFLOPS
CUDA Cores
3,584
TDP
220 W

Compute Performance

FP3236.1 TFLOPS
FP8289 TFLOPS
INT8289 TOPS
INT4583 TOPS

Architecture

MicroarchitectureRDNA
Transistors53.9 B
Compute Units56 CUs
CUDA Cores3584
Tensor CoresAI Accelerators: 112
RT CoresRay Accelerators: 56
Base Clock2070 MHz
Boost ClockUp to 2520 MHz
Sparse AccelerationSupported (Structured Sparsity)
Dynamic PrecisionSupported (FP32/FP16/FP8/INT8/INT4)

Memory & VRAM

Memory TypeGDDR6
Total Capacity16 GB
BandwidthUp to 640 GB/s
Bus Width256-bit

Power & Efficiency

TDP220 W
PSU Required650
Connectors2x8-Pin

Physical Design

Form FactorDesktop

Software Ecosystem

ROCm7.2.1
Driver StabilityAdrenalin 26.3.1 (WHQL Recommended)

System Compatibility

MotherboardBoard Type Desktop
OS CompatWindows 10 - 64-Bit Edition , Windows 11 - 64-Bit Edition , Linux x86 64-Bit

Benchmarks & Throughput

Structured Sparsity

Supported (document shows performance with structured sparsity, e.g. FP16 Matrix 289 TFLOPs vs 145 TFLOPs)

Multi-GPU Scalability

Workload Readiness

Market Authority

Key Strengths

No specific strengths can be identified.

Limitations

No limitations or trade-offs can be identified.

Expert Insight

The Radeon RX 9070 represents a powerful alternative for diversified workloads. When comparing cloud providers, consider not just the hourly rate, but also the interconnect bandwidth (InfiniBand/NVLink) and regional availability which can significantly impact total cost of ownership for large-scale training.

Glossary Terms

FP32 TFLOPS
VRAM
TDP
Cores
Information updated daily. Cloud pricing subject to vendor availability.