Hugging Face's logo

basecompute
/
Qwen3.8-27B

Text Generation
basert
apple-silicon
quantized
Model card Files
xet
 Community
3

Re-convert with the Qwen MTP head (header.speculator): basertd --speculate mtp-head

#2
by prabodbc - opened
base: refs/heads/main
←
from: refs/pr/2
Discussion Files changed
+4
-4
Files changed (2) hide show
  1. Qwen3.8-27B-Q4.base +2 -2
  2. Qwen3.8-27B-Q8.base +2 -2
Qwen3.8-27B-Q4.base CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:069f4ceca264d9e3a69ee9af725249a924024a27d46680d37da6b5851c49a294
3
- size 17467801600
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:03978674a487c567f0e5378b725b8ee1226bf554e8ecafa72943a4e7b99a0526
3
+ size 17841094656
Qwen3.8-27B-Q8.base CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6d4d362bf410c55507e894c132e03ecc0b5fbdb038b09f999ed99f59569a79d0
3
- size 28386443264
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:62cf0e6caa6270bcd3964963a259ac850c76a054e010eb19e50322fc30d8559e
3
+ size 28934225920