Comment by TD-Linux

Which is completely wrong by the way, JPEG-XL quantizes its coefficients after the DCT transform like every other lossy codec. Most codecs have at least some amount of range expansion in their DCT as well, so the values quantized might be greater bit depth than the input data.