[MPS] Float8_e4m3fn dtype not supported — works with flux1-schnell-fp8 but fails with this models
#3 opened about 1 month ago
by
listenfree
Does this safetensors has a quantization_map ?
#2 opened about 2 months ago
by
backav