Nvidia B300

Nvidia B300

Blackwell Ultra GPU for training and serving the largest foundation models.

Compare vs other GPUs →
Aggregating historical prices...

Weekly median price per GPU per hour. Nvidia B300 price history (JSON).

At a glance

Hardware

Architecture
Blackwell Ultra
Memory per GPU
288 GB HBM3e
Memory bandwidth
8,000 GB/s
Release date
Q1 2025

Market Updated 14 minutes ago

Cheapest
$6.50 / GPU / hr
on-demand · not verified in stock
Median price
$8.18 / GPU / hr
on-demand
90-day trend
Flat
median on-demand
Coverage
30 providers
11 of 131 configs in stock

B300 Pricing and Availability

30 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:

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

Search

We match the provider, its country, 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.
Runpod logo

Runpod

Our sponsor

United States of America flag USA 3 configs 1x-8x

On-Demand from $6.94 Reserved on request
Pricing On request Reservation SXM Visit website
Vast.ai logo

Vast.ai

In stock

United States of America flag USA 4 configs 2x-4x

On-Demand from $11.48 Spot from $7.50
From $11.48 / GPU / hr On-Demand Visit website
Lium logo

Lium

In stock

Distributed 1 config 1x

On-Demand from $8.50
From $8.50 / GPU / hr On-Demand SXM Visit website
Impossible Cloud logo

Impossible Cloud

In stock

Germany flag Germany 7 configs 8x

Reserved from $5.16
From $5.16 / GPU / hr Reserved (24mo) SXM Visit website
Spheron logo

Spheron

In stock

Singapore flag Singapore 1 config 1x

Spot from $5.85
From $5.85 / GPU / hr Spot SXM Visit website
Packet·ai logo

Packet·ai

In stock

United States of America flag USA 1 config 8x

Custom on request
Pricing On request Custom SXM Visit website
GPUaaS logo

GPUaaS

In stock

United States of America flag USA 3 configs 8x

Custom on request
Pricing On request Custom SXM Visit website
Canopy Wave logo

Canopy Wave

Waitlist

United States of America flag USA 1 config 8x

Custom on request
Pricing On request Custom SXM Visit website
Enverge logo

United States of America flag USA 1 config 1x

On-Demand from $7.50
From $7.50 / GPU / hr On-Demand Visit website
Fal.ai logo

United States of America flag USA 2 configs 1x

On-Demand from $8.50 Custom from $4.49
From $8.50 / GPU / hr On-Demand Visit website
Hyperstack logo

United Kingdom flag UK 1 config 1x

On-Demand from $7.40
From $7.40 / GPU / hr On-Demand Visit website
Nebius logo

Netherlands flag Netherlands 2 configs 1x

On-Demand from $7.85 Spot from $4.30
From $7.85 / GPU / hr On-Demand SXM Visit website
Bentaus logo

United States of America flag USA 2 configs 8x

On-Demand from $6.50 Reserved from $4.25
From $6.50 / GPU / hr On-Demand SXM Visit website
Gcore logo

Luxembourg flag Luxembourg 1 config 8x

On-Demand from $6.97
From $6.97 / GPU / hr On-Demand SXM Visit website
Oracle Cloud logo

United States of America flag USA 1 config 8x

On-Demand from $15.00
From $15.00 / GPU / hr On-Demand Visit website
Amazon Web Services logo

AWS

United States of America flag USA 1 config 8x

On-Demand from $17.80
From $17.80 / GPU / hr On-Demand Visit website
DigitalOcean logo

United States of America flag USA 2 configs 1x-8x

Reserved from $7.94 Spot from $11.19
From $7.94 / GPU / hr Reserved (12mo) SXM Visit website
CoreWeave logo

United States of America flag USA 1 config 8x

Spot from $4.48
From $4.48 / GPU / hr Spot SXM Visit website
TheAI Cloud logo

United Arab Emirates flag UAE 5 configs 8x

Reserved from $4.90
From $4.90 / GPU / hr Reserved (24mo) SXM Visit website
Cyfuture AI logo

India flag India 12 configs 1x-8x

Reserved from $5.51
From $5.51 / GPU / hr Reserved (12mo) Visit website
Together AI logo

United States of America flag USA 1 config 1x

Custom on request
Pricing On request Custom SXM Visit website
Exoscale logo

Switzerland flag Switzerland 1 config 1x

Custom on request
Pricing On request Custom SXM Visit website
CUDO Compute logo

United Kingdom flag UK 1 config 1x

Custom on request
Pricing On request Custom SXM Visit website
Beam logo

United States of America flag USA 1 config 8x

Custom on request
Pricing On request Custom SXM Visit website
Lyceum logo

Germany flag Germany 9 configs 1x-8x

On-Demand from $7.99 (sold out) Reserved on request Spot from $2.99 (sold out)
Pricing On request Reservation Visit website
Massed Compute logo

Massed Compute

Out of stock

United States of America flag USA 1 config 8x

On-Demand from $6.60
From $6.60 / GPU / hr On-Demand SXM Visit website
Verda logo

Verda

Out of stock

Finland flag Finland 56 configs 1x-8x

On-Demand from $7.50 Reserved from $5.62 Spot from $3.75
From $7.50 / GPU / hr On-Demand SXM Visit website
Sesterce logo

Sesterce

Out of stock

France flag France 5 configs 1x-8x

On-Demand from $8.48
From $8.48 / GPU / hr On-Demand Visit website
Scaleway logo

Scaleway

Out of stock

France flag France 1 config 8x

On-Demand from $8.71
From $8.71 / GPU / hr On-Demand SXM Visit website
Latitude.sh logo

Latitude.sh

Out of stock

Brazil flag Brazil 3 configs 8x

On-Demand from $16.00 Reserved from $5.60
From $16.00 / GPU / hr On-Demand SXM Visit website

Heads up: A provider's own page may quote a different figure, for example on tax, region or a promotion. A monthly-only plan shows a derived hourly rate. Verify before provisioning. More on how we price.

What size AI models can the B300 run?

One B300 has up to 288 GB of VRAM. In practice, that's enough memory for roughly 460B parameters at 4-bit or 125B at 16-bit, assuming a 32K context. Below are some open-weight LLMs, with the estimated memory and GPUs each one needs.

Model Parameters 4-bitINT4 / FP4 8-bitFP8 / INT8 16-bitFP16 / BF16
Alibaba Cloud logo Qwen3.8-27B
27B
1 GPU 18 GB VRAM · $5,890/mo
1 GPU 30 GB VRAM · $5,890/mo
1 GPU 58 GB VRAM · $5,890/mo
Google Cloud logo Gemma 4 31B
30.7B
1 GPU 22 GB VRAM · $5,890/mo
1 GPU 35 GB VRAM · $5,890/mo
1 GPU 69 GB VRAM · $5,890/mo
OpenAI logo GPT-OSS-120B
117B 5.1B active
1 GPU 67 GB VRAM · $5,890/mo
1 GPU 120 GB VRAM · $5,890/mo
1 GPU 237 GB VRAM · $5,890/mo
DeepSeek logo DeepSeek V4 Flash
284B 13B active
1 GPU 160 GB VRAM · $5,890/mo
2 GPUs 287 GB VRAM · $11,779/mo
4 GPUs 573 GB VRAM · $23,558/mo
Z.AI logo GLM-5.3-Flash
320B 18B active
1 GPU 178 GB VRAM · $5,890/mo
2 GPUs 322 GB VRAM · $11,779/mo
4 GPUs 642 GB VRAM · $23,558/mo
MiniMax logo MiniMax-M3
428B 23B active
1 GPU 239 GB VRAM · $5,890/mo
2 GPUs 432 GB VRAM · $11,779/mo
4 GPUs 862 GB VRAM · $23,558/mo
Moonshot AI logo Kimi K3
2.8T 104B active
8 GPUs 1,544 GB VRAM · $47,117/mo
11 GPUs 2,804 GB VRAM
22 GPUs 5,605 GB VRAM

Estimates based on the median on-demand rate. Memory is weights plus FP8 KV cache at 32K context per request (FP16/BF16 in the 16-bit column). GPU counts assume 90% of advertised VRAM is usable. No guarantee of runtime support, usable performance, or that a matching quantized build exists. How we estimate costs.

Technical Specifications

Nvidia B300 · Per GPU

Compute · dense
FP4 14,000 TFLOPS
FP8 4,500 TFLOPS 9,000 with sparsity
FP16 / BF16 2,250 TFLOPS 4,500 with sparsity
INT8 153.5 TOPS 307 with sparsity
FP32 75 TFLOPS
FP64 1.2 TFLOPS
Precision support FP4FP6FP8FP16BF16TF32FP32FP64INT8
Memory
Capacity 288 GB HBM3e
Bandwidth 8,000 GB/s
Bus width 8,192-bit
ECC Yes
Silicon
Architecture Blackwell Ultra
Process TSMC 4NP
Transistors 208 billion
Fabric and host
GPU interconnect NVLink 1,800 GB/s
Host interface PCIe 6.0 x16
Power
Board power 1,100 W
Cooling Passive
Platform
Partitioning MIG, up to 7 instances

Source: official Nvidia B300 datasheet.

Frequently Asked Questions

How much does the B300 cost per hour?

As of September 8, 2026, the median on-demand price is $8.18 per GPU per hour across 10 providers with a priced on-demand config.

Billing type Configs Median / GPU / hr Cheapest
On-demand 33 $8.18 $6.50
Reserved 70 $5.70 $4.25
Spot 18 $5.85 $4.30
Custom contract 10 $4.49

No median for custom-contract: we only show this when at least 3 providers list the GPU on that billing type. Custom-contract pricing is negotiated per deal and usually not published.

Who has the cheapest B300?

As of September 8, 2026, the lowest listed on-demand price for the B300 is $6.50 per GPU per hour from Bentaus, though we haven't verified current stock. Cheapest verified in stock: on-demand $8.50 from Lium.

How much does the B300 cost per month?

As of September 8, 2026, at 720 hours per month, one B300 costs an estimated $5,890 at the median on-demand price. Cheapest verified in stock: $6,120 per month on-demand, $3,715 reserved, $4,212 spot.

Where can I rent a B300?

GetDeploying currently tracks B300 configs from 30 providers. Lium and Vast.ai have verified in-stock on-demand configs. See the full price comparison above for every provider and config.

Are B300 prices going up or down?

As of September 8, 2026, the median on-demand price has been flat over the past 90 days, at about $8.18 per GPU per hour.

How many B300 GPUs do I need?

One B300 runs models up to roughly 460B parameters at 4-bit quantization or 125B at 16-bit, assuming a 32K context. Larger models run across multiple GPUs: the model table above shows the estimated memory and GPU count for popular open-weight LLMs.

Why choose the B300?

288GB of HBM3e and 1,800 GB/s of NVLink, a capacity matched in the Blackwell line only by the GB300. Holds 200B-parameter models on a single GPU without quantization.

When is the B300 not a good fit?

The highest cost tier in the Blackwell lineup. For most workloads, the B200 or H100 offers sufficient memory at lower cost.

Alternatives to Nvidia B300

Last updated