Hi,
is there any reference for the peak performance of INT1, INT4, INT8, INT16, INT32 for RTX3090 on Tensorcore?
Thanks!
Hi,
is there any reference for the peak performance of INT1, INT4, INT8, INT16, INT32 for RTX3090 on Tensorcore?
Thanks!
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| Blackwell Integer | 159 | 7934 | October 31, 2025 | |
| ZOTAC GAMING GeForce RTX 3070 Twin Edge INT8 inference | 3 | 2134 | November 15, 2021 | |
| Tensorrt inference runs slower in RTX4090 than RTX 3090 Ti | 3 | 2342 | January 10, 2023 | |
| Questions on RTX5090 Integer throughput | 3 | 677 | August 27, 2025 | |
| Turing Tensor core int4 operation | 3 | 3007 | December 11, 2018 | |
| Measuring T4 TensorCore Integer TOPS for roofline | 0 | 482 | November 5, 2020 | |
| cuBLAS INT8 tensor core mode vs. FP16 mode | 0 | 967 | February 15, 2019 | |
| Int8 TensorCores for Jetson | 6 | 1567 | April 26, 2023 | |
| What is official FP4:FP8 FLOPS ratio on Blackwell? | 8 | 229 | September 9, 2026 | |
| Request for official method to validate 100 INT8 TOPS | 5 | 71 | September 21, 2026 |