---
base_model:
- MiniMaxAI/MiniMax-M2.7
---
## Notes
- 04-15-2026: I've uploaded a working Q4_K_M using the findings from Unsloth regarding the blk.61.ffn_down_exps causing the `nan` issue, for the Q4_K_M I've quantized that specific tensor to Q6_K.
- 04-12-2026: The Q4_K_M I uploaded seems to have some issues, the PPL / KLD was throwing `nan` so I'll remove the model for now and try to get a working quant up tomorrow.

## Description
This repo contains specialized MoE-quants for MiniMax-M2.7. The idea being that given the huge size of the FFN tensors compared to the rest of the tensors in the model, 
it should be possible to achieve a better quality while keeping the overall size of the entire model smaller compared to a similar naive quantization. 
To that end, the quantization type default is kept in high quality and the FFN UP + FFN GATE tensors are quanted down along with the FFN DOWN tensors.

| Quant | Size | Mixture | PPL | 1-(Mean PPL(Q)/PPL(base)) | KLD |
| :--------- | :--------- | :------- | :------- | :------- | :------- |
| Q8_0 | 226.43 GiB (8.51 BPW) | Q8_0 | 7.880138 ± 0.060034 | +0.2412% | 0.029715 ± 0.000649 |
| Q5_K_M | 157.23 GiB (5.91 BPW) | Q8_0 / Q5_K / Q5_K / Q6_K | 7.871878 ± 0.059897 | +0.1361% | 0.038926 ± 0.000692 |
| Q4_K_M | 130.67 GiB (4.91 BPW) | Q8_0 / Q4_K / Q4_K / Q5_K | 7.951215 ± 0.060706 | +1.1453% | 0.059323 ± 0.000771 |
| Q4_K_S | 117.74 GiB (4.42 BPW) | Q8_0 / IQ4_XS / IQ4_XS / Q4_K | 7.968221 ± 0.060797 | +1.3616% | 0.071012 ± 0.000774 |
| IQ4_XS | 101.10 GiB (3.80 BPW) | Q8_0 / IQ3_S / IQ3_S / IQ4_XS | 8.290674 ± 0.063543 | +5.4635% | 0.128807 ± 0.001070 |
| IQ3_S | 77.86 GiB (2.92 BPW) | Q6_K / IQ2_S / IQ2_S / IQ3_S | 8.815764 ± 0.067859 | +12.1430% | 0.282740 ± 0.001687 |


![kld_graph](kld_data/01_kld_vs_filesize.png "Chart showing Pareto KLD analysis of quants")
![ppl_graph](kld_data/02_ppl_vs_filesize.png "Chart showing Pareto PPL analysis of quants")