Skip to content
Updated minutes ago
amd

Instinct MI325X

amd · cdna3 · 256 GB HBM3e · 750W TDP

VRAM

256 GB

BF16 TFLOPS

1307

Bandwidth

6000 GB/s

From

$2.49/hr

Calculate ROI with this GPU →

Spec Sheet

VRAM256 GB HBM3e
Memory Bandwidth6000 GB/s
BF16 TFLOPS1307
FP16 TFLOPS1307
FP8 TFLOPS2614
INT8 TOPS2614
TDP750W
InterconnectINFINITY-FABRIC
Max per Node8
PCIe Gen5
Tensor CoresNo

Pricing by Provider

ProviderOn-DemandReservedSpotBadge
vast_ai$3.49/hr-$2.49/hrCheapest
coreweave$4.19/hr$3.19/hr-
lambda$3.29/hr--

Compatible Models (280)

Training Capabilities

Estimated GPU count for full fine-tuning (AdamW, BF16) and QLoRA

Model SizeFull Fine-TuneQLoRA
7B model1 GPU1 GPU
13B model1 GPU1 GPU
70B model6 GPUs1 GPU

Energy Efficiency

Estimated tokens/second per Watt for popular models

Mistral 7B
1.10 t/s/WFP8
Qwen 2.5 7B
1.05 t/s/WFP8
Llama 3.1 8B
1.00 t/s/WFP8
DeepSeek V3
0.22 t/s/WFP8
Llama 3.1 70B
0.11 t/s/WFP8
Qwen 2.5 72B
0.11 t/s/WFP8

Similar GPUs

GPUVRAMBF16 TFLOPSBW (GB/s)From
Instinct MI300X192 GB13075300$1.79/hr
Groq LPU230 GB18880000$0.00/hr
B300288 GB280012000$0.00/hr
B100 SXM192 GB17508000$4.50/hr
GB200 NVL72 (per GPU)192 GB22508000$6.50/hr

Methodology Note

Performance estimates for the Instinct MI325Xare based on InferenceBench's roofline performance model with CUDA kernel-level optimization including FlashAttention v2 and PagedAttention. Memory calculations account for model weights (256 GB HBM3e available), KV-cache allocation, and activation memory. Throughput predictions use the Instinct MI325X's rated 6000 GB/s memory bandwidth and 1307 BF16 TFLOPS compute capacity as roofline ceilings, with empirical correction factors per GPU architecture (cdna3). See our full methodology.

Frequently Asked Questions

How many AI models can run on Instinct MI325X?

The Instinct MI325X can run 280 AI models from our database within a single node. Compatible models range across various parameter sizes depending on the quantization precision (BF16, FP8, INT4). Smaller models fit on a single GPU while larger models may require multi-GPU setups up to 8x Instinct MI325X.

What is the Instinct MI325X inference throughput?

The Instinct MI325X delivers 1307 BF16 TFLOPS and 2614 FP8 TFLOPS with 6000 GB/s memory bandwidth. Actual inference throughput (tokens/sec) depends on the model size, precision, and batch size. Use our calculator for model-specific throughput estimates.

How much does Instinct MI325X cost per hour?

The Instinct MI325X is available starting from $2.49/hour via vast_ai. Prices vary by provider and pricing tier (on-demand, reserved, spot). Compare pricing across all providers in the table above.