NVIDIA GeForce RTX 4080 16GB vs NVIDIA L40 48GB

The L40 is the newer part: Ada Lovelace, launched in 2023, against the GeForce RTX 4080's Ada Lovelace from 2022. Newer architectures typically add lower-precision formats and better throughput per watt, so check the precision rows your workload actually uses.

Dense throughput for the GeForce RTX 4080 against the L40: FP32 49 vs 91 TFLOPS, FP16 97 vs 181 TFLOPS, FP8 195 vs 362 TFLOPS. Sparse figures, where the vendor publishes them, appear under each dense number in the performance table.

Memory is 16 GB against 48 GB, fed at 717 GB/s versus 864 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 320 W for the GeForce RTX 4080 and 300 W for the L40. At FP32 that works out to 0.15 against 0.30 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.