You can deploy this model and start earning money today!

BennyDaBall/
MiniMax-M2.5-REAP-139B-A10B-MXFP4

Name: MiniMax-M2.5-REAP-139B-A10B-MXFP4
Rating: 2.5 (18 reviews)

Views

No views yet

MiniMax-M2.5-REAP-139B-A10B-MXFP4

Follow me on X @BennyDaBall_OG !

When you want MoE-specific compression without sending quality into the abyss.

Built from:

Quant	Size (GiB)	Notes
`MXFP4_MOE`	70.91	MoE-oriented quantization, with many non-expert tensors preserved at higher precision

This quant uses a mixed layout (as expected for MXFP4 MoE), including mxfp4, q8_0, and f32 tensors.

Use the first shard; llama.cpp resolves the rest:

llama-cli -m MiniMax-M2.5-REAP-MXFP4_MOE-00001-of-00007.gguf -ngl 0 -c 8192

MXFP4 has a different target audience than standard Q-series packs, so it gets its own clean home and simpler download surface.

You are responsible for your own use, outputs, and compliance with applicable laws and platform policies.

Follow me on X @BennyDaBall_OG !