Compute Comparison
vs
All comparisons →

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

Ada Lovelace

NVIDIA L40

wins

1

The NVIDIA L40 is the predecessor to the L40S, offering 48GB GDDR6 and Ada Lovelace architecture.

Professional inferenceRenderingVisualization AI

Performance comparison

Visual bars — winner highlighted

NVIDIA GB300 SXM6 288GB

Metric

NVIDIA L40

FP16 Tensor TFLOPS (dense)

2.8k T
181 T

Bandwidth

8.0k GB/s
864 GB/s

VRAM

288 GB
48 GB

TDP (lower=better)

1.2k W
300 W

TFLOPS/watt

2.33
0.60

GB/s per watt

6.67
2.88

Full specification table

SpecNVIDIA GB300 SXM6 288GBNVIDIA L40
ArchitectureBlackwell UltraAda Lovelace
VRAM288 GB HBM3e48 GB GDDR6
VRAM typeHBM3eGDDR6
Memory bandwidth8,000 GB/s864 GB/s
FP16 Tensor TFLOPS (dense)2,800 TFLOPS181 TFLOPS
TDP1,200 W300 W
TFLOPS/watt2.33 T/W0.60 T/W
GB/s per watt6.67 GB/s/W2.88 GB/s/W
NVLinkYesNo
Release year20252022

Live cloud pricing

On-demand pricing across providers — updated in real time

Per hourPer GB VRAM/hr

NVIDIA GB300 SXM6 288GB

No live pricing data available.

NVIDIA L40

QuantaCloud
$0.94/hr

Power efficiency analysis

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

NVIDIA GB300 SXM6 288GB

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

NVIDIA L40

FP16 TFLOPS181 TFLOPS
TDP300 W
TFLOPS/watt0.603 T/W
GB/s per watt2.88 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

NVIDIA GB300 SXM6 288GB

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

NVIDIA L40

Best cloud price$0.94/hr
TFLOPS per $193.1 T/$
GB/s per $922 GB/s/$

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

WorkloadScaleNVIDIA GB300 SXM6 288GBNVIDIA L40
7B LLM fine-tune8× GPU, 24 h—$180
70B LLM fine-tune64× GPU, 72 h—$4.3k
Inference server1× GPU, 730 h—$684
Image gen batch4× GPU, 8 h—$30

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 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

Choose NVIDIA L40 for:

  • Professional inference
  • Rendering
  • Visualization AI

Weaknesses / Limitations

  • Lower FP16 TFLOPS (181) than L40S (362) at the same VRAM and bandwidth
  • Superseded by L40S — less available on new deployments
  • GDDR6 bandwidth lower than HBM-based data center GPUs