Skip to content

GPU comparison

A100 80GB vs MI300X.

The MI300X is ahead on memory (192 GB vs 80 GB) and speed (4.2x the FP16 throughput and at least 2.5x the memory bandwidth). The A100 80GB makes sense when your model fits in 80 GB and its hourly price is lower.

Rent either on Runcrate

Live hourly price, then per GB of memory and per 100 FP16 TFLOPS.

Which is better for

Which to pick for each job. Memory fit and throughput, from the datasheets.

  • Hold a 70B model in 8-bit weights

    70 GB of 8-bit weights

    A100 80GB: 1 GPU, 10 GB spare

    MI300X: 1 GPU, 122 GB spare

    MI300X. Same GPU count, at least 2.5x the memory bandwidth.

  • Serve an 8B model in 16-bit weights

    16 GB of 16-bit weights

    A100 80GB: 1 GPU, 64 GB spare

    MI300X: 1 GPU, 176 GB spare

    MI300X. Same GPU count, at least 2.5x the memory bandwidth.

  • Fine-tune a 70B model with QLoRA

    About 46 GB with 4-bit weights

    A100 80GB: 1 GPU, 34 GB spare

    MI300X: 1 GPU, 146 GB spare

    MI300X. Same GPU count, 4.2x the FP16 throughput.

  • Train on a full node

    FP16 tensor throughput per node

    A100 80GB: 8x, 2,496 TFLOPS

    MI300X: 8x, 10,459 TFLOPS

    MI300X. 4.2x the FP16 throughput per GPU.

  • Throughput per watt

    FP16 TFLOPS per 100 W of TDP

    A100 80GB: Varies by card

    MI300X: 174 TFLOPS per 100 W

    Either. Varies by card.

Compare another pair

vs

Ampere

NVIDIA A100 80GB

from$1.73/hrCheaper
Available now · 6 regions
Memory
80 GB HBM2e
FP16 tensor
312 TFLOPS
Bandwidth
up to 2.0 TB/s
Full pricing details

CDNA 3

AMD MI300X

from$4.85/hr per GPU

on an 8-GPU machine

Available now · 1 region
Memory
192 GB HBM3
FP16 tensor
1307.4 TFLOPS
Bandwidth
5.3 TB/s
Full pricing details

At a glance

Spec by spec. The longer bar is the better number.

Memory

NVIDIA A100 80GB80 GB
AMD MI300X192 GB

Memory bandwidth

NVIDIA A100 80GBup to 2.0 TB/s
AMD MI300X5.3 TB/s

FP16 tensor

NVIDIA A100 80GB312 TFLOPS
AMD MI300X1307.4 TFLOPS

Hourly price, from (live)

NVIDIA A100 80GB$1.73/hr
AMD MI300X$4.85/hr

Full specs

The complete spec sheet.

SpecNVIDIA A100 80GBAMD MI300X
ArchitectureAmpereCDNA 3
Memory80 GB192 GB
Memory typeHBM2eHBM3
Memory bandwidthup to 2.0 TB/s5.3 TB/s
FP3219.5 TFLOPS163 TFLOPS
FP16 tensor312 TFLOPS1307.4 TFLOPS
INT8624 TOPS2610 TOPS
TDPup to 400 W750 W
GPUs per node88
Hourly price (live)from $1.73/hrfrom $4.85/hr per GPU on an 8-GPU machine

Questions

A100 80GB or MI300X. What buyers ask first.

Other comparisons

More pairs with the A100 80GB or the MI300X.