NVIDIA GeForce RTX 5090 32GB vs NVIDIA B100 SXM 192GB

The GeForce RTX 5090 is the newer part: Blackwell, launched in 2025, against the B100's Blackwell from 2024. 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 5090 against the B100: FP64 1.6 vs 30 TFLOPS, FP32 105 vs 60 TFLOPS, FP16 419 vs 1,750 TFLOPS, FP8 838 vs 3,500 TFLOPS. Sparse figures, where the vendor publishes them, appear under each dense number in the performance table.

Memory is 32 GB against 192 GB, fed at 1.8 TB/s versus 8.0 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 575 W for the GeForce RTX 5090 and 700 W for the B100. At FP32 that works out to 0.18 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.