AMD Instinct MI325X
Memory-flagship CDNA 3 accelerator with 256 GB of HBM3e — the largest single-chip memory in the inference market — and 6 TB/s of memory bandwidth. Pin-compatible with MI300X OAM platforms, positioning AMD against H200 / B100 on capacity-bound LLM workloads.
VRAM
256 GB
Memory
HBM3e
Bandwidth
6000 GB/s
TDP
1000W
Suitable Workloads
Large Language Models
Training and inference for models like GPT-4, Llama 70B+
Deep Learning Training
High-performance training for neural networks
Distributed Training
Multi-node training with fast interconnects
High-Throughput Inference
Optimized for batched inference workloads
Key Highlights
- Built on CDNA 3 architecture
- 1,216 AI accelerator cores for tensor operations
- 19,456 compute cores for parallel processing
- OAM form factor
Memory
Compute Performance
Architecture
Power & Physical
Interconnect
Pricing
Notes
FP16/BF16 figures are dense matrix throughput (no sparsity), matching the convention used for MI300X. With 2:4 sparsity, peak FP16 is 2614 TFLOPS.
Models That May Fit on AMD Instinct MI325X
Estimates based on INT8 quantization. Actual fit depends on framework and batch size.
Explore More
Added Apr 30, 2026
Last updated: Apr 30, 2026
Start building with the right model.
From model selection to production, one platform, no fragmentation.