Modders have unlocked hidden AI compute performance on NVIDIAโs RTX 20, 30 and 40-series GPUs.
The modifications remove software-level limits on certain Tensor Core operations, without changing the hardware or flashing a new VBIOS.
- RTX 4090: 165.6 โ 331.6 TFLOPS FP16 Tensor Core throughput
-RTX 3090 Ti: 87.5 โ 157.4 TFLOPS BF16 performance
- RTX 2080 Ti: 59.9 โ 111 TFLOPS FP16โFP32 performance
- RTX 3090: 73.0 โ 122.8 TFLOPS in a PyTorch GEMM test
The gains are mainly relevant to AI and compute workloads, not gaming.