Google TPU v4 32GB vs Google TPU v5e 16GB
The TPU v4 delivers 40% more BF16 throughput than the TPU v5e (275 vs 197 TFLOPS dense). The TPU v4 also carries 16 GB more memory (32 GB vs 16 GB).
Google TPU v4 32GB
Full specs →Google TPU v5e 16GB
Full specs →Specifications
Performance (TFLOPS)
FLOPS by Precision
What actually differs
The TPU v5e is the newer part: TPU, launched in 2023, against the TPU v4's TPU from 2020. Newer architectures typically add lower-precision formats and better throughput per watt, so check the precision rows your workload actually uses.
Memory is 32 GB against 16 GB, fed at 1.2 TB/s versus 800 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 TPU v4 faster than the TPU v5e?
At BF16 precision the TPU v4 reaches 275 TFLOPS dense against 197 TFLOPS for the TPU v5e. The performance table on this page lists every published precision for both GPUs.
Which has more memory, the TPU v4 or the TPU v5e?
The TPU v4 carries 32 GB of VRAM versus 16 GB for the TPU v5e. Memory bandwidth is 1.2 TB/s for the TPU v4 and 800 GB/s for the TPU v5e.
Get Comparison Updates
New GPUs added weekly. Be the first to see how they compare.