Home / GPU / MI300X
MI300X GPU Capacity Sourcing
The AMD Instinct MI300X is built on the CDNA 3 architecture, with 192GB of HBM3 memory per GPU — more per-GPU memory than NVIDIA's H100 or B200. It is typically deployed in 8-GPU platforms connected by AMD's Infinity Fabric. FindMeAGPU sources MI300X capacity from cloud and colocation providers on request.
Specifications
- Architecture
- CDNA 3
- Memory
- 192GB HBM3
- Form factor
- OAM
- Interconnect
- Infinity Fabric
Typical workloads
- LLM inference
- Training
- Fine-tuning
How sourcing works
- 1.
Tell us what you need
Submit GPU model, quantity, region, start date, and duration.
- 2.
We check across providers
We take your requirement to GPU cloud and colocation providers and check current capacity.
- 3.
You receive quotes
We contact you directly with quotes from providers that can meet your requirement.
FAQ
What is the MI300X used for?
The MI300X is used for large language model inference and training, particularly workloads that benefit from its large per-GPU memory capacity.
How does the MI300X compare to NVIDIA GPUs?
The MI300X has more memory per GPU (192GB HBM3) than the H100 or B200, which can reduce the number of GPUs needed for large models. Software tooling differs from NVIDIA's CUDA ecosystem, so it is worth checking framework compatibility for your workload.
Is MI300X capacity available now?
Availability varies by provider, region, and time and changes frequently. Submit a request and we check current capacity across our provider network.
What server configurations is the MI300X deployed in?
The MI300X is typically deployed in 8-GPU platforms connected by AMD's Infinity Fabric, similar in layout to NVIDIA's 8-GPU HGX systems.
Does FindMeAGPU sell MI300X hardware directly?
No. FindMeAGPU is a sourcing service — we take your requirement to multiple GPU cloud and colocation providers and help you find available capacity. We do not own or sell hardware.