Two Nvidia data center GPUs, one Blackwell and one Blackwell Ultra: 180 GB HBM3e against 288 GB HBM3e and memory bandwidth within 5%.
At a glance
Memory
- B200
- 180 GB
- B300
- 288 GB
1.6x more on B300
Memory bandwidth
- B200
- 7.70 TB/s
- B300
- 8.00 TB/s
Within 5%
Cheapest on-demand
- B200
- $3.75 /GPU/hr
- B300
- $6.50 /GPU/hr
42% cheaper on B200
Price comparison
Median price per GPU per hour by billing type, with the cheapest listing under it.
| Billing | B200 | B300 | Difference |
|---|---|---|---|
| On-demand | $6.70 /GPU/hr from $3.75 | $8.05 /GPU/hr from $6.50 | 17% cheaper on B200 |
| Reserved | $5.10 /GPU/hr from $3.75 | $5.70 /GPU/hr from $4.25 | 10% cheaper on B200 |
| Spot | $4.26 /GPU/hr from $3.95 | $5.71 /GPU/hr from $4.30 | 25% cheaper on B200 |
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. Medians are across providers, one vote each, so a large catalog does not outweigh a small one. Verify before provisioning. More on how we price.
Price history
B200 vs B300 price history
Weekly median on-demand price per GPU per hour
Specs comparison
| Spec | B200 | B300 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 180 GB HBM3e | 288 GB HBM3e | 1.6x more on B300 |
| Bandwidth | 7,700 GB/s | 8,000 GB/s | Within 5% |
| Bus width | 8,192-bit | ||
| Compute · dense | |||
| FP4 | 9,000 TFLOPS | 14,000 TFLOPS | 1.6x higher on B300 |
| FP8 | 4,500 TFLOPS | 4,500 TFLOPS | Identical |
| FP16 / BF16 | 2,250 TFLOPS | 2,250 TFLOPS | Identical |
| INT8 | 4,500 TOPS | 153.5 TOPS | 29x higher on B200 |
| FP32 | 75 TFLOPS | 75 TFLOPS | Identical |
| FP64 | 37 TFLOPS | 1.2 TFLOPS | 31x higher on B200 |
| Platform | |||
| Interconnect | NVLink 1,800 GB/s | NVLink 1,800 GB/s | Identical |
| Board power | 1,000 W | 1,100 W | 1.1x higher on B300 |
| Released | Q1 2024 | Q1 2025 | 12 months newer on B300 |
| Precision | |||
| Supported formats | FP4FP6FP8FP16BF16TF32FP32FP64INT8 | FP4FP6FP8FP16BF16TF32FP32FP64INT8 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: B200 datasheet and B300 datasheet.
Where to rent them
Providers listing these GPUs, those with both first, so you can switch without moving clouds.
| Provider | B200 /GPU/hr | B300 /GPU/hr |
|---|---|---|
Packet·ai
|
On-Demand $3.75 | Custom on request |
Lium
|
On-Demand $5.60 | On-Demand $8.00 |
Vast.ai
|
On-Demand $5.63 | On-Demand $9.38 |
EmpirioLabs AI
|
On-Demand $11.00 | On-Demand $16.00 |
GPUaaS
|
Custom on request | Custom on request |
GPU.ai
|
On-Demand $8.76 | -- |
Spheron
|
On-Demand $9.36 | -- |
Koyeb
|
On-Demand $5.50 | -- |
TheAI Cloud
|
-- | Reserved (24mo) $4.90 |
Impossible Cloud
|
-- | Reserved (60mo) $4.32 |
Canopy Wave
|
-- | Custom on request |
35 providers list the B200 and 32 the B300. Every listing is on the B200 page and the B300 page.
Similar GPUs
Other comparisons in the same class.
-
Nvidia H100
80 GB HBM3 · 56% less bandwidth than the B200
From $1.30 /GPU/hr Compare -
Nvidia H200
141 GB HBM3e · 38% less bandwidth than the B200
From $2.09 /GPU/hr Compare
-
Nvidia GB300
288 GB HBM3e · released Q1 2025
From $18.00 /GPU/hr Compare
Our data for Nvidia B200 was last updated on Sept. 14, 2026, and for Nvidia B300 on Sept. 14, 2026.