Why choose the GB300?

Pairs Grace CPUs with Blackwell Ultra GPUs carrying 288GB HBM3e each, connected in a 72-GPU NVLink domain. Suited for frontier-scale training and long-context reasoning inference where per-GPU memory and interconnect bandwidth are the bottleneck.

When is the GB300 not a good fit?

Rack-scale form factor with limited cloud availability at the highest cost tier. The GB200 offers a similar architecture with wider availability, and the standalone B200 is sufficient for most Blackwell workloads.

Are GB300 prices going up or down?

The median on-demand price has held steady around $15.36/hr per GPU across providers.

What size AI models can the GB300 run?

With 288GB of VRAM, the GB300 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 GB300 have?

The GB300 has 288GB of VRAM. Multi-GPU setups increase total memory, but that memory is not automatically pooled across GPUs.

What is the GB300's memory bandwidth?

The GB300 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 GB300 support?

The GB300 supports 9 precision formats. Training: BF16, FP16, TF32, FP32. Inference: FP4, FP6, FP8, INT8. Scientific: FP64.

Does the GB300 support NVLink?

Yes. The GB300 supports NVLink with 1800 GB/s of bidirectional bandwidth. This helps accelerate multi-GPU communication.

Which cloud providers offer the GB300?

The GB300 is available from 4 cloud providers, including Verda, Amazon Web Services, Gcore. Pricing and availability vary by region and billing model.

Can I rent the GB300 in the cloud?

Yes. We currently track 24 GB300 listings across 4 cloud providers: Billing type Listings Avg $/GPU/hr On-demand 3 $8.62/hr Reserved 16 $7.90/hr Spot 3 $4.31/hr Custom contract 2 Custom

Nvidia GB300

Grace CPU + Blackwell Ultra GPU superchip built for rack-scale AI reasoning and inference.

Last update 11 minutes ago

Compare vs other GPUs →

Aggregating historical prices...

Key Specifications

Architecture

Grace Blackwell Ultra

Memory

Up to 20TB HBM3e

Memory Bandwidth

576,000 GB/s

Release date

Q1 2025

GB300 Pricing and Availability

The GB300 is listed by 4 cloud providers, but none offer public on-demand pricing.

4 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.

Nebius

Netherlands 1 config 1x

Custom on request

Pricing On request Custom NVL Visit website


1x GB300 NVL72 NVIDIA GB300 NVL72	288GB	112	800GB	Custom	Custom	Custom	Not reported

Gcore

Luxembourg 1 config 72x

Custom on request

Pricing On request Custom NVL Visit website


72x GB300 NVL72	20.7TB	2592	17.3TB	Custom	Custom	Custom	Not reported

AWS

USA 1 config 72x

Reserved on request

Pricing On request Reservation Visit website


72x GB300 u-p6e-gb300x72	20.7TB	--	--	Reservation	Custom	Custom	Not reported

Verda Out of stock

Finland 21 configs 1x-4x

On-Demand from $8.62 Reserved from $6.46 Spot from $4.31

From $8.62 / GPU / hr On-Demand SXM Visit website

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 GB300?: Pairs Grace CPUs with Blackwell Ultra GPUs carrying 288GB HBM3e each, connected in a 72-GPU NVLink domain. Suited for frontier-scale training and long-context reasoning inference where per-GPU memory and interconnect bandwidth are the bottleneck.
When is the GB300 not a good fit?: Rack-scale form factor with limited cloud availability at the highest cost tier. The GB200 offers a similar architecture with wider availability, and the standalone B200 is sufficient for most Blackwell workloads.
Are GB300 prices going up or down?: The median on-demand price has held steady around $15.36/hr per GPU across providers.
What size AI models can the GB300 run?: With 288GB of VRAM, the GB300 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 GB300 have?: The GB300 has 288GB of VRAM. Multi-GPU setups increase total memory, but that memory is not automatically pooled across GPUs.
What is the GB300's memory bandwidth?: The GB300 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 GB300 support?: The GB300 supports 9 precision formats. Training: BF16, FP16, TF32, FP32. Inference: FP4, FP6, FP8, INT8. Scientific: FP64.
Does the GB300 support NVLink?: Yes. The GB300 supports NVLink with 1800 GB/s of bidirectional bandwidth. This helps accelerate multi-GPU communication.
Which cloud providers offer the GB300?: The GB300 is available from 4 cloud providers, including Verda, Amazon Web Services, Gcore. Pricing and availability vary by region and billing model.
Can I rent the GB300 in the cloud?: Yes. We currently track 24 GB300 listings across 4 cloud providers:

Billing type Listings Avg $/GPU/hr

On-demand 3 $8.62/hr

Reserved 16 $7.90/hr

Spot 3 $4.31/hr

Custom contract 2 Custom

Billing type	Listings	Avg $/GPU/hr
On-demand	3	$8.62/hr
Reserved	16	$7.90/hr
Spot	3	$4.31/hr
Custom contract	2	Custom

Technical Specifications


Configuration	36 Grace CPUs : 72 Blackwell Ultra GPUs
FP4 Tensor Core	1,440 PFLOPS¹ \| 1,080 PFLOPS
FP8/FP6 Tensor Core¹	720 PFLOPS
INT8 Tensor Core¹	24 POPS
FP16/BF16 Tensor Core¹	360 PFLOPS
TF32 Tensor Core¹	180 PFLOPS
FP32	6 PFLOPS
FP64	100 TFLOPS
GPU Memory \| Bandwidth	Up to 20 TB HBM3e \| 576 TB/s
NVLink Bandwidth	130 TB/s
CPU Core Count	2,592 Arm® Neoverse V2 cores
CPU Memory \| Bandwidth	Up to 17 TB LPDDR5X \| Up to 14.3 TB/s