Hugging Face's logo

mradermacher
/
Gelato-30B-A3B-i1-GGUF

Transformers
GGUF
English
imatrix
conversational
Model card Files
xet
Community
mradermacher commited on
Commit
532f8c7
·
verified ·
1 Parent(s): 0133ce4

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +10 -0
README.md CHANGED
@@ -42,8 +42,18 @@ more details, including on how to concatenate multi-part files.
42
  | Link | Type | Size/GB | Notes |
43
  |:-----|:-----|--------:|:------|
44
  | [GGUF](https://huggingface.co/mradermacher/Gelato-30B-A3B-i1-GGUF/resolve/main/Gelato-30B-A3B.imatrix.gguf) | imatrix | 0.2 | imatrix file (for creating your own qwuants) |
 
 
 
45
  | [GGUF](https://huggingface.co/mradermacher/Gelato-30B-A3B-i1-GGUF/resolve/main/Gelato-30B-A3B.i1-Q2_K.gguf) | i1-Q2_K | 11.4 | IQ3_XXS probably better |
 
 
46
  | [GGUF](https://huggingface.co/mradermacher/Gelato-30B-A3B-i1-GGUF/resolve/main/Gelato-30B-A3B.i1-IQ3_M.gguf) | i1-IQ3_M | 13.6 | |
 
 
 
 
 
47
 
48
  Here is a handy graph by ikawrakow comparing some lower-quality quant
49
  types (lower is better):
 
42
  | Link | Type | Size/GB | Notes |
43
  |:-----|:-----|--------:|:------|
44
  | [GGUF](https://huggingface.co/mradermacher/Gelato-30B-A3B-i1-GGUF/resolve/main/Gelato-30B-A3B.imatrix.gguf) | imatrix | 0.2 | imatrix file (for creating your own qwuants) |
45
+ | [GGUF](https://huggingface.co/mradermacher/Gelato-30B-A3B-i1-GGUF/resolve/main/Gelato-30B-A3B.i1-IQ1_M.gguf) | i1-IQ1_M | 7.2 | mostly desperate |
46
+ | [GGUF](https://huggingface.co/mradermacher/Gelato-30B-A3B-i1-GGUF/resolve/main/Gelato-30B-A3B.i1-IQ2_M.gguf) | i1-IQ2_M | 10.3 | |
47
+ | [GGUF](https://huggingface.co/mradermacher/Gelato-30B-A3B-i1-GGUF/resolve/main/Gelato-30B-A3B.i1-Q2_K_S.gguf) | i1-Q2_K_S | 10.6 | very low quality |
48
  | [GGUF](https://huggingface.co/mradermacher/Gelato-30B-A3B-i1-GGUF/resolve/main/Gelato-30B-A3B.i1-Q2_K.gguf) | i1-Q2_K | 11.4 | IQ3_XXS probably better |
49
+ | [GGUF](https://huggingface.co/mradermacher/Gelato-30B-A3B-i1-GGUF/resolve/main/Gelato-30B-A3B.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 11.9 | lower quality |
50
+ | [GGUF](https://huggingface.co/mradermacher/Gelato-30B-A3B-i1-GGUF/resolve/main/Gelato-30B-A3B.i1-Q3_K_S.gguf) | i1-Q3_K_S | 13.4 | IQ3_XS probably better |
51
  | [GGUF](https://huggingface.co/mradermacher/Gelato-30B-A3B-i1-GGUF/resolve/main/Gelato-30B-A3B.i1-IQ3_M.gguf) | i1-IQ3_M | 13.6 | |
52
+ | [GGUF](https://huggingface.co/mradermacher/Gelato-30B-A3B-i1-GGUF/resolve/main/Gelato-30B-A3B.i1-Q3_K_M.gguf) | i1-Q3_K_M | 14.8 | IQ3_S probably better |
53
+ | [GGUF](https://huggingface.co/mradermacher/Gelato-30B-A3B-i1-GGUF/resolve/main/Gelato-30B-A3B.i1-IQ4_XS.gguf) | i1-IQ4_XS | 16.5 | |
54
+ | [GGUF](https://huggingface.co/mradermacher/Gelato-30B-A3B-i1-GGUF/resolve/main/Gelato-30B-A3B.i1-Q4_K_S.gguf) | i1-Q4_K_S | 17.6 | optimal size/speed/quality |
55
+ | [GGUF](https://huggingface.co/mradermacher/Gelato-30B-A3B-i1-GGUF/resolve/main/Gelato-30B-A3B.i1-Q4_K_M.gguf) | i1-Q4_K_M | 18.7 | fast, recommended |
56
+ | [GGUF](https://huggingface.co/mradermacher/Gelato-30B-A3B-i1-GGUF/resolve/main/Gelato-30B-A3B.i1-Q6_K.gguf) | i1-Q6_K | 25.2 | practically like static Q6_K |
57
 
58
  Here is a handy graph by ikawrakow comparing some lower-quality quant
59
  types (lower is better):