Apple M2 Ultra vs Apple M2 Max

The M2 Ultra delivers 2.0x the FP32 throughput of the M2 Max (27 vs 14 TFLOPS dense). The M2 Ultra also carries 96 GB more memory (192 GB vs 96 GB).

M2 Ultra: Apple M2, 2023 M2 Max: Apple M2, 2023
FP32 dense lead
2.0x
M2 Ultra: 27 vs 14 TFLOPS
VRAM
192 vs 96 GB
96 GB more for the M2 Ultra
Bandwidth
800 GB/s vs 400 GB/s
M2 Ultra moves data faster

Specifications

M2 Ultra
M2 Max
Architecture
Apple M2Apple M2
Launch Year
20232023
Form Factor
SoCSoC
VRAM
192 GB96 GB
Memory Bandwidth
800 GB/s400 GB/s
TDP
Process Node
5nm5nm

Performance (TFLOPS)

M2 Ultra
M2 Max
FP32
27.2 TFlops 13.6 TFlops

FLOPS by Precision

What actually differs

Both GPUs launched in 2023: the M2 Ultra on Apple M2 and the M2 Max on Apple M2.

Dense throughput for the M2 Ultra against the M2 Max: FP32 27 vs 14 TFLOPS. Sparse figures, where the vendor publishes them, appear under each dense number in the performance table.

Memory is 192 GB against 96 GB, fed at 800 GB/s versus 400 GB/s. More memory per GPU means larger models fit before you have to shard across cards, which often matters more than raw TFLOPS for inference.

For LLM training and inference, weight the FP8 and FP16 rows and memory capacity most heavily. For scientific and HPC workloads, the FP64 row is the one to read.

Frequently asked questions

Is the M2 Ultra faster than the M2 Max?

At FP32 precision the M2 Ultra reaches 27 TFLOPS dense against 14 TFLOPS for the M2 Max. The performance table on this page lists every published precision for both GPUs.

Which has more memory, the M2 Ultra or the M2 Max?

The M2 Ultra carries 192 GB of VRAM versus 96 GB for the M2 Max. Memory bandwidth is 800 GB/s for the M2 Ultra and 400 GB/s for the M2 Max.

Get Comparison Updates

New GPUs added weekly. Be the first to see how they compare.