exllamav3 quantizations of MiniMaxAI/MiniMax-M2.5.
2.00 bpw h6 61.054 GiB
3.00 bpw h6 81.613 GiB
4.00 bpw h6 108.087 GiB
5.00 bpw h6 134.561 GiB (Uploading)
measurement.json - 2.0bpw_H6 vs 3.0bpw_H6
measurement.json - 3.0bpw_H6 vs 4.0bpw_H6
Model tree for MikeRoz/MiniMax-M2.5-exl3
Base model
MiniMaxAI/MiniMax-M2.5