⚡ Live Pricing

Rent NVIDIA B200 192GB: All Cloud Providers & Prices (April 2026)

Real-time NVIDIA B200 cloud pricing from 6 providers. Cheapest on-demand: $5.29/hr (Lambda). Updated daily by GridStackHub.

Last updated: 2026-04-24 — 6 pricing records
Cheapest On-Demand
$5.29/hr
Lambda
Cheapest Reserved (1yr)
N/A
Not widely available
Providers Available
6
Active 2026-04-24

NVIDIA B200 Cloud Pricing — All Providers

Sorted by cheapest per-GPU hourly rate. Includes on-demand, spot, and reserved pricing where available.

Provider Price/hr Type Region VRAM Updated
LambdaLowest
1x B200 SXM
$5.29/hr On-Demand US 192 GB Sun Apr 12 2026 00:00:00 GMT+0000 (Coordinated Universal Time)
CoreWeave
B200 SXM (Early Access)
$5.49/hr On-Demand US 192 GB Sun Apr 12 2026 00:00:00 GMT+0000 (Coordinated Universal Time)
RunPod
NVIDIA B200
$5.98/hr On-Demand us-east-1 180 GB Wed Apr 22 2026 00:00:00 GMT+0000 (Coordinated Universal Time)
Google Cloud
a4-highgpu-8g (8x B200)
$6.60/hr
$52.80/hr for 8× node
On-Demand us-central1 192 GB Sun Apr 12 2026 00:00:00 GMT+0000 (Coordinated Universal Time)
AWS
p6.48xlarge (8x B200)
$6.90/hr
$55.20/hr for 8× node
On-Demand us-east-1 192 GB Sun Apr 12 2026 00:00:00 GMT+0000 (Coordinated Universal Time)
Azure
ND B200 v6 (8x B200)
$7.05/hr
$56.40/hr for 8× node
On-Demand East US 192 GB Sun Apr 12 2026 00:00:00 GMT+0000 (Coordinated Universal Time)

NVIDIA B200 192GB Specifications

Key hardware specifications for the NVIDIA B200.

ArchitectureBlackwell (SXM)
VRAM192GB HBM3e
Memory Bandwidth8.0 TB/s
FP8 Throughput9,000 TFLOPS
FP16 Throughput4,500 TFLOPS
GPU-to-GPU Bandwidth1.8 TB/s (NVLink 5)
TDP1,000W
Gen5th Gen NVLink

About the NVIDIA B200

The NVIDIA B200 is the current flagship GPU, built on NVIDIA's Blackwell architecture. With 192GB HBM3e memory and 9,000 TFLOPS FP8 compute, it delivers 2.27× the throughput of an H200 and 2.27× that of an H100 SXM5 in memory-bandwidth-bound workloads.

B200 supply remains constrained in April 2026. Lambda ($5.29/hr) and CoreWeave ($5.49/hr) are the most accessible on-demand providers. Hyperscaler pricing — AWS at $6.90/GPU, Google Cloud at $6.60/GPU, Azure at $7.05/GPU — reflects normalized per-GPU rates from 8-GPU nodes.

The 192GB HBM3e enables running frontier-size models on a single GPU. For inference on 70B–405B parameter models, B200 on-demand eliminates multi-GPU tensor parallelism in many configurations, simplifying deployment and reducing interconnect overhead.

NVLink 5 interconnect on B200 nodes provides 1.8 TB/s GPU-to-GPU bandwidth — double H100 SXM5 NVSwitch bandwidth. For dense training on 1T+ parameter models, the B200's bandwidth advantage compounds across nodes. CoreWeave and Lambda have the deepest current B200 inventory.

Related GPU Pages

Frequently Asked Questions

What is the cheapest B200 cloud provider?+
The cheapest NVIDIA B200 cloud rental as of April 2026 is Lambda at $5.29/hr per GPU (1x B200 SXM, 192GB HBM3e, on-demand). CoreWeave is second at $5.49/hr for B200 SXM early-access capacity. RunPod lists B200 on-demand at $5.98/hr. For 8-GPU nodes, Google Cloud is cheapest at $52.80/hr ($6.60/GPU), followed by AWS at $55.20/hr ($6.90/GPU) and Azure at $56.40/hr ($7.05/GPU).
How much does a B200 GPU cost per month?+
At $5.29/hr on Lambda, running one B200 24/7 for 30 days costs approximately $3,809/month. CoreWeave at $5.49/hr runs $3,953/month. For 8-GPU nodes on AWS at $55.20/hr, monthly cost is approximately $39,744. Reserved/committed pricing is not yet widely available for B200 — pricing is expected to decline as supply expands through 2026.
B200 vs H100 SXM5: which should I rent?+
B200 delivers 2.27× higher FP8 throughput (9,000 vs 3,958 TFLOPS) and 2.39× more memory bandwidth (8.0 vs 3.35 TB/s). At current pricing, B200 costs roughly 2.7× more than H100 SXM5 per GPU-hour. For memory-bandwidth-bound inference on large models (70B+), B200 is cost-neutral or better. For smaller workloads (7B–13B), H100 SXM5 is better value.
Is B200 available on spot/preemptible pricing?+
As of April 2026, B200 spot/preemptible pricing is not widely available. Supply is constrained enough that providers maintain on-demand pricing without a spot tier. Expect spot pricing to appear at Vast.ai and RunPod community as supply expands through mid-2026. Set a price alert on GridStackHub to be notified when B200 spot pricing goes live.
What workloads are best suited for B200?+
B200 is optimal for: (1) Large-scale LLM inference on 70B–405B parameter models, where 192GB VRAM enables single-GPU serving. (2) Frontier model training where FP8 throughput and NVLink 5 bandwidth reduce wall-clock training time. (3) MoE architectures with sparse activation patterns benefiting from high memory bandwidth. For inference on smaller models (7B–30B), H100 SXM5 or H200 offer better cost efficiency.

Compare All Providers for Your Workload

Use GridStackHub's GPU cost calculator to get a ranked comparison with hidden-cost breakdown (egress + storage) across all providers.

📊 Open Calculator View All GPU Pricing