AMD MI300X vs L40S: Specs, VRAM & Price

Short answer: choose the AMD MI300X when 192 GB of VRAM is the gating requirement. Choose the L40S when 48 GB is enough for the workload and the lower hourly rate points that way.

The AMD MI300X has 192 GB of VRAM; the L40S has 48 GB.

You can rent both by the hour, no purchase required. AMD MI300X from $3.29/GPU/hr; L40S from $1.03/GPU/hr. The L40S is 69% cheaper at the entry rate.

AMD MI300X

Good for training and serving frontier-scale LLMs

$3.29/hr
Lowest / GPU
$3.29/hr
Median / GPU
192 GB
VRAM
CDNA3
Architecture

L40S

Good for fine-tuning mid-size models and high-throughput inference

$1.03/hr
Lowest / GPU
$2.47/hr
Median / GPU
48 GB
VRAM
Ada Lovelace
Architecture

Last updated: 2026-10-16 UTCRefreshes every 6 hours

Specs side by side

SpecAMD MI300XL40S
VRAM192 GB48 GB
Memory bandwidth5.3 TB/s864 GB/s
ArchitectureCDNA3Ada Lovelace
Compute capabilityn/a8.9
Precisions in hardwareFP32, FP16, BF16, FP8, INT4FP32, FP16, BF16, FP8, INT4
Peak FP16/BF16 tensor TFLOPS1,307.4 TFLOPS362 TFLOPS
Peak FP8 tensor TFLOPS2,614.9 TFLOPS733 TFLOPS
NVLink / multi-GPU interconnectInfinity Fabric–
Max board power (TDP)750W350W
Form factor–PCIe
Regions14

On-demand price

RateAMD MI300XL40S
Lowest per GPU/hr$3.29/GPU/hr$1.07/GPU/hr

Which is better for AI: AMD MI300X or L40S?

Choose the AMD MI300X when 192 GB of VRAM is the gating requirement. Choose the L40S when 48 GB is enough for the workload and the lower hourly rate points that way.

AMD MI300X vs L40S: common questions

Is the AMD MI300X or the L40S better for AI?

Short answer: choose the AMD MI300X when 192 GB of VRAM is the gating requirement. Choose the L40S when 48 GB is enough for the workload and the lower hourly rate points that way.

Is the AMD MI300X or the L40S cheaper to rent?

On Aquanode's live marketplace the AMD MI300X starts at $3.29/GPU/hr (median $3.29/GPU/hr) and the L40S starts at $1.03/GPU/hr (median $2.47/GPU/hr). The L40S is the cheaper of the two at the entry rate, by 69%.

What is the difference between the AMD MI300X and the L40S?

The AMD MI300X has 192 GB of VRAM against the L40S's 48 GB; the AMD MI300X is a CDNA3 part and the L40S is Ada Lovelace; both run BF16, FP8, INT4 workloads in hardware. On price, the AMD MI300X lists from $3.29/GPU/hr and the L40S from $1.03/GPU/hr.

How widely available are the AMD MI300X and the L40S?

The AMD MI300X is listed in 1 region and the L40S in 4 regions right now.

How much does 1,000 GPU-hours cost on the AMD MI300X vs the L40S?

At the lowest rates listed today, 1,000 GPU-hours costs $3,289 on the AMD MI300X and $1,028 on the L40S, a difference of $2,261 for the same runtime. Rates are per GPU per hour and update hourly.

How much VRAM does the AMD MI300X have compared to the L40S?

The AMD MI300X has 192 GB of VRAM; the L40S has 48 GB.

Does the AMD MI300X or the L40S support NVLink?

The AMD MI300X supports Infinity Fabric; the L40S has no dedicated GPU-to-GPU interconnect.

How this comparison is calculated

Every price is normalized to a per-GPU hourly rate using the same pipeline as every other pricing surface on Aquanode, and only the cheapest qualifying offer per provider is shown. A dash means that provider does not currently list that GPU.

Architecture, compute capability and supported precisions come from each vendor's published datasheet for that generation, not from the marketplace feed. This page regenerates at most once per hour.

Memory bandwidth, tensor TFLOPS, NVLink/interconnect and max board power are the same verified-from-datasheet figures shown on each model's own AMD MI300X page and L40S page, which link the primary vendor source for that model. A dash here means Aquanode hasn't verified that figure from a primary source yet, never a guess.

Go further

Check what fits in each card's memory with the AMD MI300X VRAM calculator and the L40S VRAM calculator, see vendor-published throughput in the GPU benchmarks and specs table, or browse every model by generation in the GPU index.

Submit the job. Everything after that is ours.

Sign up in 60 seconds. Pay for the GPU minutes you actually use.

© 2026 Aquanode. All rights reserved.

All trademarks, logos and brand names are the property of their respective owners.