MI300X vs MI325X
The MI300X delivers 192GB at 5.3TB/s, while the MI325X offers 256GB at 6TB/s. Choose MI300X for current availability, MI325X for larger memory.
Overview
|
AMD MI300X
|
AMD MI325X
|
|
|---|---|---|
| Summary | AMD's flagship data center GPU for large-model inference. | Highest memory capacity AMD GPU for single-chip model serving. |
| Architecture | CDNA 3 | CDNA3 |
| Memory | 192GB HBM3 | 256GB HBM3E |
| Memory Bandwidth | 5,300 GB/s | 6,000 GB/s |
| Performance | 1,307 TFLOPS (FP16) | 2,615 TFLOPS (FP16) |
| Release Date | Q4 2023 | Q4 2024 |
| Performance Segment | High Performance | High Performance |
| Datasheet | Link | Link |
Price comparison
MI300X Pricing
Listings for the MI300X reach $7.86/hr, often reflecting a premium for high availability. However, you might be able to find available instances from as low as $0.95/hr per GPU (spot instance).
How we sort results
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 "View all offerings" to access the full list with sortable columns and filters for billing type, GPU count, vCPUs, and RAM.
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 | Billing | Total/h | |
|---|---|---|---|---|
DigitalOcean
|
1x | On-Demand Pay-as-you-go pricing. No term commitments. | $1.99 | View |
Hot Aisle
|
1x | On-Demand Pay-as-you-go pricing. No term commitments. | $1.99 | View |
|
Hot Aisle
|
1x | On-Demand Pay-as-you-go pricing. No term commitments. | $1.99 | View |
Crusoe
|
8x | On-Demand Pay-as-you-go pricing. No term commitments. | $27.60 | View |
|
Hot Aisle
|
2x | On-Demand Pay-as-you-go pricing. No term commitments. | $3.98 | View |
|
Hot Aisle
|
4x | On-Demand Pay-as-you-go pricing. No term commitments. | $7.96 | View |
Vultr
|
8x | Reserved (24mo) Reserved capacity with a 24-month minimum commitment. | $14.80 | View |
Azure
|
8x | Spot Variable spot pricing with potential interruptions. | $11.62 | View |
|
Crusoe
|
8x | Spot Variable spot pricing with potential interruptions. | $7.60 | View |
Cirrascale
|
8x | Reserved (12mo) 12-month contract: $17,999/month | $24.99 | View |
Oracle Cloud
|
8x | On-Demand Pay-as-you-go pricing. No term commitments. | $48.00 | View |
|
DigitalOcean
|
8x | Reserved (12mo) Reserved capacity with a 1-year minimum commitment. | $15.04 | View |
|
Cirrascale
|
8x | Reserved (6mo) 6-month contract: $20,249/month | $28.12 | View |
|
Azure
|
8x | On-Demand Pay-as-you-go pricing. No term commitments. | $62.85 | View |
|
Cirrascale
|
8x | Reserved (3mo) 3-month contract: $21,379/month | $29.69 | View |
| No offerings matching your filters. | ||||
MI325X Pricing
With a spread of only 11%, price optimization yields diminishing returns for the MI325X. The market is relatively flat, with the lowest rate sitting at $2.00/hr per GPU (36-month reservation).
How we sort results
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 "View all offerings" to access the full list with sortable columns and filters for billing type, GPU count, vCPUs, and RAM.
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 | Billing | Total/h | |
|---|---|---|---|---|
|
Vultr
|
8x | Reserved (36mo) Reserved capacity with a 3-year minimum commitment. | $16.00 | View |
|
DigitalOcean
|
8x | Reserved (12mo) Reserved capacity with a 1-year minimum commitment. | $16.80 | View |
TensorWave
|
8x | Custom Custom terms negotiated with the provider. | $18.00 | View |
| No offerings matching your filters. | ||||
Heads up: We do our best to keep these 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 with the provider before provisioning.
Alternative GPUs
Here's how these GPUs compare against alternatives:
AMD MI300X
vs
Nvidia H100
AMD MI300X
vs
Nvidia A100
AMD MI325X
vs
Nvidia H200
AMD MI325X
vs
Nvidia B200
Our data for AMD MI300X was last updated on March 25, 2026, and for AMD MI325X on March 25, 2026.