NVFP4 uses 16-element blocks rather than standard 32-element blocks.
@jackcookjack new toy for you
This block quant visualizer is another page in our LLM Engineer's Almanac -- a one-stop shop for engineers looking to own their inference.
It joins another page for intuition-building with fp quants, focused on the binary representations.
https://modal.com/llm-almanac/quant-formats/e4::0x4c
Users are enthusiastic about the visualizer demonstrating NVFP4 MXFP4 block quantization effects, calling it a new toy.
No Digg Deeper questions have been answered for this story yet.
Low-precision floats are weird. I have been building up my intuition by playing with them outside of inference/training.
@AAAzzam and I cooked up this visualizer for micro-scaling/block quant formats like NVFP4, MXFP4, and friends.
Try it: https://modal.com/llm-almanac/block-quants/nvidia-fp4
This block quant visualizer is another page in our LLM Engineer's Almanac -- a one-stop shop for engineers looking to own their inference.
It joins another page for intuition-building with fp quants, focused on the binary representations.
https://modal.com/llm-almanac/quant-formats/e4::0x4c
Low-precision floats are weird. I have been building up my intuition by playing with them outside of inference/training.
@AAAzzam and I cooked up this visualizer for micro-scaling/block quant formats like NVFP4, MXFP4, and friends.
Try it: https://modal.com/llm-almanac/block-quants/nvidia-fp4
@charles_irl @AAAzzam dont listen to the haters king you look amazing in NVFP4
Low-precision floats are weird. I have been building up my intuition by playing with them outside of inference/training.
@AAAzzam and I cooked up this visualizer for micro-scaling/block quant formats like NVFP4, MXFP4, and friends.
Try it: https://modal.com/llm-almanac/block-quants/nvidia-fp4