Why choose the GB200?

Combined Grace CPU + Blackwell GPU with 192GB HBM3e and 1800 GB/s NVLink. Unified memory architecture eliminates CPU-GPU bottlenecks for large-scale AI training.

When is the GB200 not a good fit?

Specialized superchip form factor with limited cloud availability. The standalone B200 is more widely offered and sufficient for most Blackwell workloads.

Are GB200 prices going up or down?

The median on-demand price across providers has risen about 42% since August 2025, from $13.25 to $18.77/hr per GPU. This reflects the market-wide median, which moves both when providers change prices and when lower or higher priced offerings enter the market.

What size AI models can the GB200 run?

With 192GB of VRAM, the GB200 can usually run 70B-class models with headroom, and may handle much larger models in 4-bit quantized form depending on runtime overhead, KV cache, and context length.

How much VRAM does the GB200 have?

The GB200 has 192GB of VRAM. Multi-GPU setups increase total memory, but that memory is not automatically pooled across GPUs.

What is the GB200's memory bandwidth?

The GB200 has 8,000 GB/s of memory bandwidth. Higher bandwidth helps with faster data transfer between GPU memory and compute cores.

What data types does the GB200 support?

The GB200 supports 9 precision formats. Training: BF16, FP16, TF32, FP32. Inference: FP4, FP6, FP8, INT8. Scientific: FP64.

Does the GB200 support NVLink?

Yes. The GB200 supports NVLink with 1800 GB/s of bidirectional bandwidth. This helps accelerate multi-GPU communication.

How much does the GB200 cost per hour?

GB200 pricing currently ranges from $10.50/hr to $27.04/hr per GPU, depending on the provider, instance type, and billing model.

How much does the GB200 cost per month?

At 720 hours per month, one GB200 can cost between $7,560.00 to $19,468.80 per month, depending on the provider. Reserved and spot pricing can lower that further.

Which cloud providers offer the GB200?

The GB200 is available from 8 cloud providers, including Amazon Web Services, CoreWeave, Nebius. Pricing and availability vary by region and billing model.

Can I rent the GB200 in the cloud?

Yes. We currently track 9 GB200 listings across 8 cloud providers: Billing type Listings Avg $/GPU/hr On-demand 3 $17.85/hr Reserved 1 $10.58/hr Spot 1 $27.04/hr Custom contract 4 Custom

Nvidia GB200

Grace CPU + Blackwell GPU superchip with unified memory architecture.

Last update 2 minutes ago

Compare vs other GPUs →

Aggregating historical prices...

Key Specifications

Architecture

Grace Hopper Blackwell

Memory

Up to 13.4TB HBM3e

Memory Bandwidth

576,000 GB/s

Release date

Q1 2024

GB200 Pricing and Availability

We're tracking 8 cloud providers for the GB200, and the pricing spread is significant. While the average sits at $17.85/hr, the lowest price is currently $10.50/hr per GPU (on-demand).

8 providers

How this list works

The two views

By provider shows one card per company, placed where its best offer ranked. By configuration lists every offer.

Order

What you can rent comes first: in stock, then waitlist, then not reported, then out of stock. Priced offers rank ahead of quote-only, and on-demand ahead of other billing types.

Within a group, five factors set the order, largest first:

Location: datacenter proximity, blended with provider HQ.
Price: hourly price, per GPU and in total.
Billing type: reserved ahead of spot, spot ahead of quote-only.
Specs: more VRAM, vCPUs and RAM.
Provider diversity: in By configuration, a provider's repeat rows rank slightly lower, so one company can't take all the top spots.

Search

We match the provider, GPU model, form factor, billing type, availability and instance name. Matching is partial and case-insensitive. Everyday words work too, so "interruptible" finds spot and "sold out" finds out of stock.

Transparency and funding

Ads and sponsors: Paid placements sit at the top of the list and are always labeled as sponsored content. Sponsorship never influences the organic ranking itself.
Affiliates: Affiliate links are marked. We may earn a commission if you click them, but commissions never affect the order.
Prices: Shown in USD, converted at daily reference rates where a provider publishes in another currency. A month means 720 hours. How we estimate costs.

CoreWeave In stock

USA 1 config 4x

On-Demand from $10.50

From $10.50 / GPU / hr On-Demand NVL Visit website


4x GB200 NVL72	744GB	144	960GB	On-Demand	$10.50	$42.00	In stock

Azure In stock

USA 2 configs 4x

On-Demand from $27.04 Spot from $27.04

From $27.04 / GPU / hr On-Demand Visit website


4x GB200	768GB	128	900GB	Spot	$27.04	$108.16	In stock
4x GB200	768GB	128	900GB	On-Demand	$27.04	$108.16	In stock

Oracle Cloud

USA 1 config 4x

On-Demand from $16.00

From $16.00 / GPU / hr On-Demand NVL Visit website


4x GB200 NVL72 BM.GPU.GB200.41	756GB	--	--	On-Demand	$16.00	$64.00	Not reported

AWS

USA 1 config 72x

Reserved from $10.58

From $10.58 / GPU / hr Reservation Visit website


72x GB200 u-p6e-gb200x72	13.4TB	2592	17.3TB	Reservation	$10.58	$761.90	Not reported

Together

USA 1 config 72x

Custom on request

Pricing On request Custom NVL Visit website


72x GB200 NVL	13.4TB	--	--	Custom	Custom	Custom	Not reported

Canopy Wave

USA 1 config 4x

Custom on request

Pricing On request Custom NVL Visit website


4x GB200 NVL72	768GB	144	960GB	Custom	Custom	Custom	Not reported

CUDO

UK 1 config 1x

Custom on request

Pricing On request Custom NVL Visit website


1x GB200 NVL GB200 NVL72	192GB	--	--	Custom	Custom	Custom	Not reported

Nebius

Netherlands 1 config 1x

Custom on request

Pricing On request Custom NVL Visit website


1x GB200 NVL72 NVIDIA GB200 NVL72	192GB	112	800GB	Custom	Custom	Custom	Not reported

Heads up: We do our best to keep these specs & prices accurate. However, cloud costs may fluctuate based on region, usage, and other factors not listed here. These are estimates based on common setups and are for informational purposes only. Always verify current rates & exact specs with the provider before provisioning.

Frequently Asked Questions

Why choose the GB200?: Combined Grace CPU + Blackwell GPU with 192GB HBM3e and 1800 GB/s NVLink. Unified memory architecture eliminates CPU-GPU bottlenecks for large-scale AI training.
When is the GB200 not a good fit?: Specialized superchip form factor with limited cloud availability. The standalone B200 is more widely offered and sufficient for most Blackwell workloads.
Are GB200 prices going up or down?: The median on-demand price across providers has risen about 42% since August 2025, from $13.25 to $18.77/hr per GPU. This reflects the market-wide median, which moves both when providers change prices and when lower or higher priced offerings enter the market.
What size AI models can the GB200 run?: With 192GB of VRAM, the GB200 can usually run 70B-class models with headroom, and may handle much larger models in 4-bit quantized form depending on runtime overhead, KV cache, and context length.
How much VRAM does the GB200 have?: The GB200 has 192GB of VRAM. Multi-GPU setups increase total memory, but that memory is not automatically pooled across GPUs.
What is the GB200's memory bandwidth?: The GB200 has 8,000 GB/s of memory bandwidth. Higher bandwidth helps with faster data transfer between GPU memory and compute cores.
What data types does the GB200 support?: The GB200 supports 9 precision formats. Training: BF16, FP16, TF32, FP32. Inference: FP4, FP6, FP8, INT8. Scientific: FP64.
Does the GB200 support NVLink?: Yes. The GB200 supports NVLink with 1800 GB/s of bidirectional bandwidth. This helps accelerate multi-GPU communication.
How much does the GB200 cost per hour?: GB200 pricing currently ranges from $10.50/hr to $27.04/hr per GPU, depending on the provider, instance type, and billing model.
How much does the GB200 cost per month?: At 720 hours per month, one GB200 can cost between $7,560.00 to $19,468.80 per month, depending on the provider. Reserved and spot pricing can lower that further.
Which cloud providers offer the GB200?: The GB200 is available from 8 cloud providers, including Amazon Web Services, CoreWeave, Nebius. Pricing and availability vary by region and billing model.
Can I rent the GB200 in the cloud?: Yes. We currently track 9 GB200 listings across 8 cloud providers:

Billing type Listings Avg $/GPU/hr

On-demand 3 $17.85/hr

Reserved 1 $10.58/hr

Spot 1 $27.04/hr

Custom contract 4 Custom

Billing type	Listings	Avg $/GPU/hr
On-demand	3	$17.85/hr
Reserved	1	$10.58/hr
Spot	1	$27.04/hr
Custom contract	4	Custom

Technical Specifications


Configuration	36 Grace CPU : 72 Blackwell GPUs
FP4 Tensor Core¹	1,440 PFLOPS
FP8/FP6 Tensor Core¹	720 PFLOPS
INT8 Tensor Core¹	720 POPS
FP16/BF16 Tensor Core¹	360 PFLOPS
TF32 Tensor Core	180 PFLOPS
FP32	5,760 TFLOPS
FP64	2,880 TFLOPS
FP64 Tensor Core	2,880 TFLOPS
GPU Memory \| Bandwidth	Up to 13.4 TB HBM3e \| 576 TB/s
NVLink Bandwidth	130TB/s
CPU Core Count	2,592 Arm® Neoverse V2 cores
CPU Memory \| Bandwidth	Up to 17 TB LPDDR5X \| Up to 18.4 TB/s