PlanetGPU
Search/B300

B300

288GB 1300W 3 providers 37 offers

Blackwell Ultra (B300) with 288 GB HBM3e for the largest models and longest context windows.

Cheapest
$1.880/hr
Vast.ai
Median
$15.000/hr
across 37 offers
P90
$60.000/hr
90th percentile
Prices updated 1 hour ago3 providers trackedHow we collect prices
Memory
288 GB
TDP
1300 W
Bandwidth
8000 GB/s
FP16
1100 TFLOPS
Brand
NVIDIA
Series
NVIDIA
Year
2025
Arch
Blackwell Ultra

Price by provider

cheapest rate · $/GPU/hr · lower is better
Vast.ai
0 regions
$1.880/hr
Verda
0 regions
$2.625/hr
AWS
0 regions
$33.546/hr

3 providers

ProviderPrice RangeMedianTypeCPU / RAMAction
VA
Vast.ai
Best price
$1.880$55.406
$10.679 Spot
32-256
48-3096GB
Browse
VE
Verda
$2.625$60.000
$12.750 Spot
30-240
255-2040GB
Browse
AW
AWS
$33.546$170.900
$142.416 Spot
192-192
4096-4096GB
Browse
Best Deal Right Now
VA
Vast.ai
$1.880/hr
per GPU / hour
Configurationx1 · B300
System RAM48-3096GB
Region—
Launch on Vast.ai

Estimated Running Cost

At cheapest rate · per GPU

1 hour$1.88
24 hours$45.12
1 week$315.84
1 month$32937.60

Available At

VA
Vast.ai
$1.880/hr
VE
Verda
$2.625/hr
AW
AWS
$33.546/hr

Compare

See how B300 stacks up against other GPUs.

B300 vs H100

Explore NVIDIA B300

Related comparisons, provider pages, raw specs, and model-fit tools.

NVIDIA B300 vs NVIDIA H100Compare live rental prices, VRAM, FP16 throughput, and provider availability.NVIDIA B300 vs NVIDIA H200Compare live rental prices, VRAM, FP16 throughput, and provider availability.NVIDIA B300 vs NVIDIA A100Compare live rental prices, VRAM, FP16 throughput, and provider availability.NVIDIA B300 on Vast.aiSee SKUs, regions, spot status, and current prices from $1.880/hr.NVIDIA B300 on VerdaSee SKUs, regions, spot status, and current prices from $2.625/hr.NVIDIA B300 on AWSSee SKUs, regions, spot status, and current prices from $33.546/hr.NVIDIA B300 full specificationsOpen the raw GPU Ark spec detail page for clocks, memory, cores, and API fields.Find models that fit this class of GPUMatch LLM size and precision to GPUs with enough VRAM.