Testing Nvidia's FP4: Running 70B LLMs on a Single RTX 5090 with Real Benchmarks(ai.gopubby.com)
ai.gopubby.com
Testing Nvidia's FP4: Running 70B LLMs on a Single RTX 5090 with Real Benchmarks
https://ai.gopubby.com/fp4-quantization-nvfp4-blackwell-tutorial-13dfc854ed0c
0 comments
—