The 2080 Ti comes out on top as the best GPU in 2018 for training neural nets.
Results summary
As of October 8, 2018, the NVIDIA RTX 2080 Ti is the best GPU for deep learning research on a single GPU system running TensorFlow. A typical single GPU system with this GPU will be:
- 37% faster than the
1080 Tiwith FP32, 62% faster with FP16 - 35% faster than the
2080with FP32, 47% faster with FP16 - 96% as fast as the
Titan Vwith FP32, 3% faster with FP16 - 80% as fast as the
Tesla V100with FP32, 82% as fast with FP16
Note that all experiments utilized Tensor Cores when available and are priced out on a complete single GPU system cost. As a system builder and AI research company, we’re trying to make benchmarks that are scientific, reproducible, correlate with real world training scenarios, and have accurate prices.
https://lambdalabs.com/blog/best-gpu-tensorflow-2080-ti-vs-v100-vs-titan-v-vs-1080-ti-benchmark/