rzgar/minimax_h3_ref2va_fp8_e4m3fn
Quantization vs MiniMaxAI/MiniMax-H3. rzgar/minimax_h3_ref2va_fp8_e4m3fn is a quantized version of the other (F16/F8_E4M3).
Quantization vs MiniMaxAI/MiniMax-H3. rzgar/minimax_h3_ref2va_fp8_e4m3fn is a quantized version of the other (F16/F8_E4M3).