Skip to content
Updated minutes ago
amd

Instinct MI210

amd · cdna2 · 64 GB HBM2e · 300W TDP

VRAM

64 GB

BF16 TFLOPS

181

Bandwidth

1638 GB/s

From

$0.99/hr

Calculate ROI with this GPU →

Spec Sheet

VRAM64 GB HBM2e
Memory Bandwidth1638 GB/s
BF16 TFLOPS181
FP16 TFLOPS181
FP8 TFLOPS181
INT8 TOPS181
TDP300W
InterconnectINFINITY-FABRIC
Max per Node4
PCIe Gen4
Tensor CoresNo

Pricing by Provider

ProviderOn-DemandReservedSpotBadge
lambda$0.99/hr--Cheapest

Compatible Models (260)

Training Capabilities

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

Model SizeFull Fine-TuneQLoRA
7B model3 GPUs1 GPU
13B model4 GPUs1 GPU
70B model21 GPUs1 GPU

Energy Efficiency

Estimated tokens/second per Watt for popular models

Mistral 7B
0.75 t/s/WFP8
Qwen 2.5 7B
0.72 t/s/WFP8
Llama 3.1 8B
0.68 t/s/WFP8
Llama 3.1 70B
0.08 t/s/WFP8
Qwen 2.5 72B
0.08 t/s/WFP8

Similar GPUs

GPUVRAMBF16 TFLOPSBW (GB/s)From
Instinct MI250X128 GB3833277$0.79/hr
A1664 GB16.8232$0.72/hr
H100 SXM80 GB9903350$1.89/hr
H100 PCIe80 GB7562000$1.89/hr
A100 80GB SXM80 GB3122039$1.19/hr

Methodology Note

Performance estimates for the Instinct MI210are based on InferenceBench's roofline performance model with CUDA kernel-level optimization including FlashAttention v2 and PagedAttention. Memory calculations account for model weights (64 GB HBM2e available), KV-cache allocation, and activation memory. Throughput predictions use the Instinct MI210's rated 1638 GB/s memory bandwidth and 181 BF16 TFLOPS compute capacity as roofline ceilings, with empirical correction factors per GPU architecture (cdna2). See our full methodology.

Frequently Asked Questions

How many AI models can run on Instinct MI210?

The Instinct MI210 can run 260 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 4x Instinct MI210.

What is the Instinct MI210 inference throughput?

The Instinct MI210 delivers 181 BF16 TFLOPS and 181 FP8 TFLOPS with 1638 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 MI210 cost per hour?

The Instinct MI210 is available starting from $0.99/hour via lambda. Prices vary by provider and pricing tier (on-demand, reserved, spot). Compare pricing across all providers in the table above.