Hugging Face's logo

turboderp
/
Qwen3.6-27B-MTP-exl3

Model card Files
xet
 Community
turboderp commited on
Commit
1dfaf05
·
verified ·
1 Parent(s): 1f28d17

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +3 -3
README.md CHANGED
@@ -1,8 +1,8 @@
1
  These are the MTP tensors for Qwen3.6-27B, quantized at 4 bpw.
2
 
3
- Checkpoints quantized with exllamav3 v0.0.41+ will automatically contain MTP layers.
4
- For quantized checkpoints from earlier versions, place this file in the model directory
5
- to enable MTP drafting.
6
 
7
  Quick benchmark (RTX 6000 Pro, 4.00 bpw target model):
8
 
 
1
  These are the MTP tensors for Qwen3.6-27B, quantized at 4 bpw.
2
 
3
+ Qwen3.5/3.6 checkpoints quantized with exllamav3 v0.0.41+ will automatically contain MTP
4
+ layers. For quantized checkpoints from earlier versions, place this file in the model
5
+ directory to enable MTP drafting.
6
 
7
  Quick benchmark (RTX 6000 Pro, 4.00 bpw target model):
8