Intel Data Center GPU Max 1100 48GB vs Intel Data Center GPU Max 1550 128GB

The Data Center GPU Max 1550 delivers 2.0x the FP16 throughput of the Data Center GPU Max 1100 (839 vs 420 TFLOPS dense). The Data Center GPU Max 1550 also carries 80 GB more memory (128 GB vs 48 GB).

Data Center GPU Max 1100: Xe-HPC, 2023 Data Center GPU Max 1550: Xe-HPC, 2023
FP16 dense lead
2.0x
Data Center GPU Max 1550: 839 vs 420 TFLOPS
VRAM
48 vs 128 GB
80 GB more for the Data Center GPU Max 1550
Bandwidth
1.2 TB/s vs 3.3 TB/s
Data Center GPU Max 1550 moves data faster
TDP
300 W vs 600 W
Data Center GPU Max 1100 draws 300 W less

Specifications

Data Center GPU Max 1100
Data Center GPU Max 1550
Architecture
Xe-HPCXe-HPC
Launch Year
20232023
Form Factor
PCIeOAM
VRAM
48 GB128 GB
Memory Bandwidth
1.2 TB/s3.3 TB/s
TDP
300 W600 W
Process Node

Performance (TFLOPS)

Data Center GPU Max 1100
Data Center GPU Max 1550
FP64
22.2 TFlops 52.0 TFlops
FP32
22.2 TFlops 52.0 TFlops
FP16
419.5 TFlops 839.0 TFlops
BF16
419.5 TFlops 839.0 TFlops
INT8
839.0 TFlops 1678.0 TFlops

FLOPS by Precision

What actually differs

Both GPUs launched in 2023: the Data Center GPU Max 1100 on Xe-HPC and the Data Center GPU Max 1550 on Xe-HPC.

Dense throughput for the Data Center GPU Max 1100 against the Data Center GPU Max 1550: FP64 22 vs 52 TFLOPS, FP32 22 vs 52 TFLOPS, FP16 420 vs 839 TFLOPS. Sparse figures, where the vendor publishes them, appear under each dense number in the performance table.

Memory is 48 GB against 128 GB, fed at 1.2 TB/s versus 3.3 TB/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.

Power budgets are 300 W for the Data Center GPU Max 1100 and 600 W for the Data Center GPU Max 1550. At FP32 that works out to 0.07 against 0.09 TFLOPS per watt.

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 Data Center GPU Max 1100 faster than the Data Center GPU Max 1550?

At FP16 precision the Data Center GPU Max 1550 reaches 839 TFLOPS dense against 420 TFLOPS for the Data Center GPU Max 1100. The performance table on this page lists every published precision for both GPUs.

Which has more memory, the Data Center GPU Max 1100 or the Data Center GPU Max 1550?

The Data Center GPU Max 1550 carries 128 GB of VRAM versus 48 GB for the Data Center GPU Max 1100. Memory bandwidth is 1.2 TB/s for the Data Center GPU Max 1100 and 3.3 TB/s for the Data Center GPU Max 1550.

How much power do the Data Center GPU Max 1100 and the Data Center GPU Max 1550 draw?

The Data Center GPU Max 1100 is rated at 300 W TDP and the Data Center GPU Max 1550 at 600 W. On FP32 throughput per watt, the Data Center GPU Max 1550 is the more efficient part.

Get Comparison Updates

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