Comes with 288GB HBM3e - the highest VRAM in the Blackwell lineup. 1800 GB/s NVLink. Can serve 200B+ parameter models on a single GPU without quantization.

When is the B300 not a good fit?

For most workloads, the B200 or H100 might offer sufficient memory at lower cost.

Are B300 prices going up or down?

On-demand pricing has increased by about 51% since November 2025, from $5.00 to $7.55/hr per GPU.

What size AI models can the B300 run?

With 288GB of VRAM, the B300 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 B300 have?

The B300 has 288GB of VRAM. Multi-GPU setups increase total memory, but that memory is not automatically pooled across GPUs.

What is the B300's memory bandwidth?

The B300 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 B300 support?

The B300 supports 9 precision formats. Training: BF16, FP16, TF32, FP32. Inference: FP4, FP6, FP8, INT8. Scientific: FP64.

Does the B300 support NVLink?

Yes. The B300 supports NVLink with 1800 GB/s of bidirectional bandwidth. This helps accelerate multi-GPU communication.

How much does the B300 cost per hour?

B300 pricing currently ranges from $3.13/hr to $18.00/hr per GPU, depending on the provider, instance type, and billing model.

How much does the B300 cost per month?

At 720 hours per month, one B300 can cost between $2,250.20 to $12,960.00 per month, depending on the provider. Reserved and spot pricing can lower that further.

Which cloud providers offer the B300?

The B300 is available from 13 cloud providers, including Verda, Nebius, Enverge. Pricing and availability vary by region and billing model.

Can I rent the B300 in the cloud?

Yes. We currently track 35 B300 listings across 13 cloud providers: Billing type Listings Avg $/GPU/hr On-demand 19 $8.62/hr Reserved 6 $5.85/hr Spot 8 $4.80/hr Custom contract 2 Custom

Nvidia B300

Highest VRAM in the Blackwell lineup for training the largest foundation models.

Compare vs other GPUs →

Aggregating historical prices...

Key Specifications

Architecture

Blackwell Ultra

Memory

Up to 288 HBM3e

Memory Bandwidth

8,000 GB/s

Release date

Q4 2025

Compare Cloud Provider Prices

Listings for the B300 reach $18.00/hr, often reflecting a premium for high availability. However, you might be able to find available instances from as low as $3.13/hr per GPU (spot instance).

Billing

Max price/hr

Min GPUs

Min vCPUs

Min RAM (GB)

of shown

Default ranking

Our algorithm weighs five factors to find the most relevant matches for you:

Price: We blend total hourly price with price-per-GPU to balance affordability and value.
Specs: We favor offers with higher CPU, RAM, and GPU Memory.
Billing: We favor on-demand billing for simplicity and flexibility over spot instances, reservations, and custom quotes.
Location: We blend both datacenter proximity and provider HQ location. Datacenter location matters for latency, while HQ location matters for compliance and support.
Provider diversity: Each time a provider appears in the list, their subsequent offerings are ranked a little lower, so one provider's offerings don't crowd out the top positions.

Sorting and filtering

Click any column header to sort by that column. Use the filters above the table to narrow results by billing type, GPU count, vCPUs, or RAM. Custom sorting resets the default relevance ranking.

Transparency and funding

Ads and sponsors: Any paid placements are fixed at the top and clearly labeled as sponsored content.

Affiliates: Any affiliate links will be indicated to you as well. We may earn a commission if you click them, but this never influences the ranking order.

Provider	GPUs	Total VRAM	vCPUs	RAM	Billing	$/GPU/h	Total/h	Availability
Vast.ai Our sponsor	2x B300 2x B300 275GB (41798315)	550GB	192	1.0TB	Spot Variable spot pricing with potential interruptions.	$5.44	$10.88	Available	View
Nebius	1x B300 SXM5 1x B300 288GB SXM5 (NVIDIA HGX B300)	288GB	24	346GB	On-Demand Pay-as-you-go pricing. No term commitments.	$7.85	$7.85	Available	View
Sesterce	1x B300 1x B300 288GB	288GB	30	255GB	On-Demand Pay-as-you-go pricing. No term commitments.	$8.70	$8.70	Available	View
Lyceum	1x B300 1x B300 288GB (B300)	288GB	32	240GB	On-Demand Per GPU, per hour. Instances configurable from 1 to 8 GPUs with per-second billing.	$7.99	$7.99	Unknown	View
Verda	1x B300 NVL 1x B300 268GB NVL (1B300.30V)	268GB	30	255GB	On-Demand Pay-as-you-go pricing. No term commitments.	$7.50	$7.50	Unknown	View
Enverge	1x B300 1x B300 288GB	288GB	--	--	On-Demand Pay-as-you-go pricing. No term commitments.	$7.50	$7.50	Unknown	View
Verda	2x B300 NVL 2x B300 268GB NVL (2B300.60V)	536GB	60	510GB	On-Demand Pay-as-you-go pricing. No term commitments.	$7.50	$15.00	Unknown	View
Vast.ai	4x B300 4x B300 275GB (38294439)	1.1TB	192	1.0TB	On-Demand Pay-as-you-go pricing. No term commitments.	$3.75	$15.00	Available	View
Runcrate	1x B300 NVL 1x B300 288GB NVL	288GB	30	255GB	On-Demand Pay-as-you-go pricing. No term commitments.	$8.48	$8.48	Available	View
Oracle Cloud	4x B300 NVL 4x B300 279GB NVL (BM.GPU.GB300.4)	1.1TB	--	--	On-Demand Pay-as-you-go pricing. No term commitments.	$18.00	$72.00	Unknown	View
CoreWeave	8x B300 SXM 8x B300 270GB SXM	2.2TB	192	4.1TB	Spot Variable spot pricing with potential interruptions.	$4.48	$35.84	Available	View
DigitalOcean	8x B300 8x B300 288GB	2.3TB	--	--	Reserved (12mo) Reserved capacity with a 1-year minimum commitment.	$5.65	$45.20	Unknown	View
Sesterce	8x B300 8x B300 288GB	2.3TB	240	2.0TB	On-Demand Pay-as-you-go pricing. No term commitments.	$8.48	$67.80	Available	View
Impossible Cloud	8x B300 SXM 8x B300 288GB SXM (nvidia-b300-8gpu-sxm-36mo-3576)	2.3TB	--	--	Reserved (36mo) Reserved capacity with a 3-year minimum commitment.	$4.47	$35.76	Waitlist	View
Vast.ai	2x B300 2x B300 275GB (41798310)	550GB	192	1.0TB	On-Demand Pay-as-you-go pricing. No term commitments.	$6.07	$12.13	Available	View
Nebius	1x B300 SXM5 1x B300 288GB SXM5 (NVIDIA HGX B300)	288GB	24	346GB	Spot Variable spot pricing with potential interruptions.	$4.30	$4.30	Available	View
Verda	4x B300 NVL 4x B300 268GB NVL (4B300.120V)	1.1TB	120	1.0TB	On-Demand Pay-as-you-go pricing. No term commitments.	$7.50	$30.00	Unknown	View
Impossible Cloud	8x B300 SXM 8x B300 288GB SXM (nvidia-b300-8gpu-sxm-12mo-3824)	2.3TB	--	--	Reserved (12mo) Reserved capacity with a 1-year minimum commitment.	$4.78	$38.24	Waitlist	View
AWS	8x B300 8x B300 268GB (p6-b300.48xlarge)	2.1TB	192	4.1TB	On-Demand Pay-as-you-go pricing. No term commitments.	$17.80	$142.42	Available	View
Vast.ai	8x B300 8x B300 275GB (38294437)	2.2TB	192	2.1TB	On-Demand Pay-as-you-go pricing. No term commitments.	$3.75	$30.00	Available	View
No offerings matching your filters.

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 B300?: Comes with 288GB HBM3e - the highest VRAM in the Blackwell lineup. 1800 GB/s NVLink. Can serve 200B+ parameter models on a single GPU without quantization.
When is the B300 not a good fit?: For most workloads, the B200 or H100 might offer sufficient memory at lower cost.
Are B300 prices going up or down?: On-demand pricing has increased by about 51% since November 2025, from $5.00 to $7.55/hr per GPU.
What size AI models can the B300 run?: With 288GB of VRAM, the B300 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 B300 have?: The B300 has 288GB of VRAM. Multi-GPU setups increase total memory, but that memory is not automatically pooled across GPUs.
What is the B300's memory bandwidth?: The B300 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 B300 support?: The B300 supports 9 precision formats. Training: BF16, FP16, TF32, FP32. Inference: FP4, FP6, FP8, INT8. Scientific: FP64.
Does the B300 support NVLink?: Yes. The B300 supports NVLink with 1800 GB/s of bidirectional bandwidth. This helps accelerate multi-GPU communication.
How much does the B300 cost per hour?: B300 pricing currently ranges from $3.13/hr to $18.00/hr per GPU, depending on the provider, instance type, and billing model.
How much does the B300 cost per month?: At 720 hours per month, one B300 can cost between $2,250.20 to $12,960.00 per month, depending on the provider. Reserved and spot pricing can lower that further.
Which cloud providers offer the B300?: The B300 is available from 13 cloud providers, including Verda, Nebius, Enverge. Pricing and availability vary by region and billing model.
Can I rent the B300 in the cloud?: Yes. We currently track 35 B300 listings across 13 cloud providers:

Billing type Listings Avg $/GPU/hr

On-demand 19 $8.62/hr

Reserved 6 $5.85/hr

Spot 8 $4.80/hr

Custom contract 2 Custom

Billing type	Listings	Avg $/GPU/hr
On-demand	19	$8.62/hr
Reserved	6	$5.85/hr
Spot	8	$4.80/hr
Custom contract	2	Custom

Technical Specifications

Feature	Specification
GPU	NVIDIA Blackwell Ultra GPUs
Total GPU Memory	2.3TB
Performance	144 PFLOPS FP4 inference, 72 PFLOPS FP8 training
NVIDIA NVLink Switch System	2x
NVIDIA NVLink Bandwidth	14.4 TB/s aggregate bandwidth
Power Consumption	~14kW
CPU	Intel® Xeon® 6776P Processors
Networking	8x OSFP ports (800Gb/s InfiniBand/Ethernet), 2x dual-port QSFP112 BlueField-3 DPU (400Gb/s InfiniBand/Ethernet)
Management Network	1GbE onboard NIC with RJ45, 1GbE RJ45 Host BMC
Storage	OS: 2x 1.9TB NVMe M.2, Internal: 8x 3.84TB NVMe E1.S
Software	NVIDIA DGX OS, Mission Control, Base Command Manager, AI Enterprise. Supports RHEL, Rocky, Ubuntu.
Rack Units	10RU
Operating Temperature	10°C - 35°C