B300
NVIDIA · Blackwell Ultra
288 GBHBM3e
8 TB/sBandwidth
4.5 PFFP8 dense
$4,899 /GPU/mo$6,552 · −25%
AMD · CDNA 4 · OAM · 8-GPU UBB
Rent the AMD Instinct MI350X by the month.
The AMD Instinct MI350X is the air-cooled sibling of the MI355X: the same 288 GB of HBM3E at 8 TB/s, 4.6 PF of dense FP8 and 9.2 PF of FP4 at 1,000 W. It is the most GPU memory per dollar in our whole catalog.
288 GBHBM3E per GPU
8 TB/sMemory bandwidth
4.6 PFFP8 dense
9.2 PFFP4 dense
Why the MI350X
What the MI350X is best at
- 288 GB HBM3E. The same memory capacity as MI355X and B300.
- 4.6 PF FP8 dense. About 92% of MI355X throughput at a lower monthly price.
- Infinity Fabric, 1.08 TB/s. Eight GPUs fully connected for tensor parallelism.
- Long-context inference
- Large-batch fine-tuning
- Serving several mid-size models on one node
Allocations of MI350X are small: when it reads “low stock”, there are only a few servers left.
Pricing
MI350X monthly price by server size
One flat price per GPU, the same from 1 to 8 GPUs. The market median is the public on-demand list price of specialist GPU clouds × 730 hours.
| Server | CoreGPU / month | Market median / month | You save / month | You save / year | ≈ per GPU-hour |
|---|---|---|---|---|---|
| 1× MI350X | $2,249 | $773 | $9,276 | $3.08 | |
| 2× MI350X | $4,498 | $1,546 | $18,552 | $3.08 | |
| 4× MI350X | $8,996 | $3,092 | $37,104 | $3.08 | |
| 8× MI350X | $17,992 | $6,184 | $74,208 | $3.08 |
Prices in USD, excluding taxes. Market median checked Nov 4, 2026 across specialist GPU clouds; hyperscalers excluded. All GPU prices
Compare
MI350X vs MI355X
Same memory, about 8% less compute. Choose MI350X to save on the monthly price when you are memory-bound rather than compute-bound.
| MI350X | MI355X | |
|---|---|---|
| Architecture | CDNA 4 | CDNA 4 |
| GPU memory | 288 GB HBM3E | 288 GB HBM3E |
| Memory bandwidth | 8 TB/s | 8 TB/s |
| FP8 dense | 4.6 PF | 5 PF |
| FP4 dense | 9.2 PF | 10.1 PF |
| Interconnect | Infinity Fabric · 1.08 TB/s | Infinity Fabric · 1.08 TB/s |
| Power (TDP) | 1,000 W | 1,400 W |
| Price per GPU / month | $2,249 | $2,449 |
| Price per GB of memory | $7.81 | $8.50 |
| Price per FP8 PFLOPS | $489 | $490 |
| Availability | Only 6 left | 21 available |
Specifications
AMD Instinct MI350X technical specs
Per-GPU figures from the vendor datasheet. Tensor throughput is dense (without sparsity).
| Model | AMD Instinct MI350X |
|---|---|
| Architecture | CDNA 4 |
| Form factor | OAM · 8-GPU UBB |
| GPU memory | 288 GB HBM3E |
| Memory bandwidth | 8 TB/s |
| FP8 Tensor (dense) | 4.6 PF |
| FP4 Tensor (dense) | 9.2 PF |
| Interconnect | Infinity Fabric · 1.08 TB/s |
| Power (TDP) | 1,000 W |
| GPUs per server | 1 to 8 |
| Availability | Only 6 left |
FAQ
MI350X rental: common questions
Another question? Our engineers answer within one business day.
Contact usHow much does it cost to rent an AMD Instinct MI350X per month?
$2,249 per GPU per month at CoreGPU, about $3.08 per GPU-hour, versus a market median of $3,022 ($4.14/hour × 730 h). A full 8-GPU server is $17,992 per month, taxes excluded.
Is the MI350X available right now?
Yes: only 6 left in the current allocation of 48 GPUs. In-stock servers are delivered with root SSH access in under 60 seconds after payment.
How many MI350X GPUs can I rent?
From 1 to 8 GPUs per server, at the same price per GPU. Servers with 8 GPUs come with the full Infinity Fabric mesh. Need more? Rent several servers or ask for a private allocation.
MI350X or MI355X?
Same memory, about 8% less compute. Choose MI350X to save on the monthly price when you are memory-bound rather than compute-bound. See the full MI350X vs MI355X comparison.
What do I get with a MI350X server?
Dedicated GPUs (never shared or sliced), root SSH access, your choice of a clean OS or a ready image with ROCm, PyTorch and vLLM, local NVMe scratch, and month-to-month billing with no setup fee.
Alternatives
Other GPUs in stock now
Deploy your MI350X server in under 60 seconds
Pick 1 to 8 GPUs in the configurator above. Month to month, cancel anytime.