NVIDIA A40 48GB vs NVIDIA GeForce RTX 3080 Ti 12GB

Both GPUs launched in 2021: the A40 on Ampere and the GeForce RTX 3080 Ti on Ampere.

Dense throughput for the A40 against the GeForce RTX 3080 Ti: FP32 37 vs 34 TFLOPS, FP16 150 vs 34 TFLOPS. Sparse figures, where the vendor publishes them, appear under each dense number in the performance table.

Memory is 48 GB against 12 GB, fed at 696 GB/s versus 912 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.

Power budgets are 300 W for the A40 and 350 W for the GeForce RTX 3080 Ti. At FP32 that works out to 0.12 against 0.10 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.