Hugging Face's logo

Green-Sky
/
bark-ggml

ggml
bark.cpp
quantized
16-bit
Model card Files Community
Green-Sky commited on
Commit
52e4260
•
1 Parent(s): b09e386

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +2 -6
README.md CHANGED
@@ -4,9 +4,7 @@ tags:
4
  - ggml
5
  - bark.cpp
6
  - quantized
7
- - 8-bit
8
  - 16-bit
9
- - 32-bit
10
  base_model: suno/bark
11
  quantized_by: Green-Sky
12
  ---
@@ -15,8 +13,6 @@ quantized_by: Green-Sky
15
 
16
  This is the large variant https://huggingface.co/suno/bark
17
 
18
- done using bark.cpp commit [d22ad71](https://github.com/PABannier/bark.cpp/commit/d22ad710e5c6ea62595877e5635e30f9c40442bc)
19
 
20
- I only uploaded f32, f16 and q8_0 quants, because q8_0 already changes the result significantly.
21
-
22
- Don't use the f32 version of encodec, it does not work with bark right now.
 
4
  - ggml
5
  - bark.cpp
6
  - quantized
 
7
  - 16-bit
 
8
  base_model: suno/bark
9
  quantized_by: Green-Sky
10
  ---
 
13
 
14
  This is the large variant https://huggingface.co/suno/bark
15
 
16
+ done using bark.cpp commit [964467c](https://github.com/PABannier/bark.cpp/commit/964467c9f752805a6f3777f1cc9bc7a665e79208)
17
 
18
+ I only uploaded f16, since f32 and other quantized models currently dont work.