Compute Comparison
vs
All comparisons →

CDNA 3

AMD Instinct MI300X

wins

1

The AMD MI300X leads all GPUs on memory capacity (192GB HBM3) and bandwidth (5.3 TB/s), making it exceptional for serving very large language models that don't fit on H100.

Memory-bound LLM inferenceLarge model servingHPC

Blackwell Ultra

NVIDIA GB300 SXM6 288GB

wins

4

The GB300 SXM6 (Grace Blackwell Ultra) is a CPU+GPU superchip with 288 GB HBM3e and 8 TB/s memory bandwidth.

Frontier AI trainingInference at scaleMemory-bound LLM serving

Performance comparison

Visual bars — winner highlighted

AMD Instinct MI300X

Metric

NVIDIA GB300 SXM6 288GB

FP16 Tensor TFLOPS (dense)

1.3k T
2.8k T

Bandwidth

5.3k GB/s
8.0k GB/s

VRAM

192 GB
288 GB

TDP (lower=better)

750 W
1.2k W

TFLOPS/watt

1.74
2.33

GB/s per watt

7.07
6.67

Full specification table

SpecAMD Instinct MI300XNVIDIA GB300 SXM6 288GB
ArchitectureCDNA 3Blackwell Ultra
VRAM192 GB HBM3288 GB HBM3e
VRAM typeHBM3HBM3e
Memory bandwidth5,300 GB/s8,000 GB/s
FP16 Tensor TFLOPS (dense)1,307 TFLOPS2,800 TFLOPS
TDP750 W1,200 W
TFLOPS/watt1.74 T/W2.33 T/W
GB/s per watt7.07 GB/s/W6.67 GB/s/W
NVLinkNoYes
Release year20232025

Live cloud pricing

On-demand pricing across providers — updated in real time

Per hourPer GB VRAM/hr

AMD Instinct MI300X

FluidStack
$31.50/hr
TensorWave
$3.56/hr
DigitalOcean
$1.99/hr
Vultr
$1.85/hr
TensorDock
$3.56/hr
RunPod
$2.39/hr
Crusoe Cloud
$3.45/hr

NVIDIA GB300 SXM6 288GB

No live pricing data available.

Power efficiency analysis

TFLOPS/watt and GB/s/watt — critical for data center TCO

AMD Instinct MI300X

FP16 TFLOPS1,307 TFLOPS
TDP750 W
TFLOPS/watt1.743 T/W
GB/s per watt7.07 GB/s/W

NVIDIA GB300 SXM6 288GB

FP16 TFLOPS2,800 TFLOPS
TDP1,200 W
TFLOPS/watt2.333 T/W
GB/s per watt6.67 GB/s/W

TFLOPS/watt measures compute efficiency — how much AI throughput you get per watt of power consumed. For data centers with PUE of 1.2–1.5, a 10% improvement in TFLOPS/watt translates directly to lower electricity costs and cooling requirements. GB/s/watt measures memory bandwidth efficiency, which is the binding constraint for memory-bound LLM inference workloads.

Performance-per-dollar

TFLOPS/$ and GB/s/$ at current best cloud prices

AMD Instinct MI300X

Best cloud price$1.85/hr
TFLOPS per $706.5 T/$
GB/s per $2865 GB/s/$

NVIDIA GB300 SXM6 288GB

Best cloud priceN/A
TFLOPS per $—
GB/s per $—

TFLOPS/$ = FP16 TFLOPS ÷ best on-demand $/hr. Higher is better — more compute per dollar. GB/s/$ = memory bandwidth ÷ best on-demand $/hr. For memory-bound inference workloads (most LLM serving), GB/s/$ is often a better predictor of cost-efficiency than TFLOPS/$. Prices are live from the Compute Comparison API and update in real time.

Workload Cost Comparison

Total job cost = best on-demand $/hr × GPU count × hours

WorkloadScaleAMD Instinct MI300XNVIDIA GB300 SXM6 288GB
7B LLM fine-tune8× GPU, 24 h$355—
70B LLM fine-tune64× GPU, 72 h$8.5k—
Inference server1× GPU, 730 h$1.4k—
Image gen batch4× GPU, 8 h$59—

Based on current best on-demand prices. Spot pricing typically 40–70% lower. Actual cost depends on provider, region, and availability.

When to choose each GPU

Choose AMD Instinct MI300X for:

  • Memory-bound LLM inference
  • Large model serving
  • HPC

Weaknesses / Limitations

  • ROCm software ecosystem is less mature than CUDA — some frameworks require extra porting effort
  • Fewer cloud providers offer MI300X compared to H100
  • 750W TDP — high power draw comparable to H100 SXM

Choose NVIDIA GB300 SXM6 288GB for:

  • Frontier AI training
  • Inference at scale
  • Memory-bound LLM serving

Weaknesses / Limitations

  • Very limited cloud availability — still ramping supply
  • Extremely high cost — enterprise/hyperscaler pricing only
  • Requires specialized NVLink-C2C infrastructure