noctrex/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill

I have created a normal one, and also an importance-aware MXFP4\MOE quantization that dynamically allocates precision based on tensor importance scores from an imatrix I created with codetiny.
This is a coding optimized quantization and is slightly larger than the mainline MXFP4\_MOE, and the way it works is that it keeps a better quantization depending on the importance of each tensor.

!Quantization Types

• BF16 (16-bit) for highly important tensors (>75% importance)
• Q8_0 (8-bit) for moderately important tensors (>60% importance)
• MXFP4 (4-bit) for less important tensors (<50% importance)

!Quantization per Layer Count

As I've mentioned it is experimental, and still not have done any benchmark on it, to see if it's any better than mainline, but you are freely to try it out and report back!