Hugging Face's logo

mradermacher
/
Cathallama-70B-i1-GGUF

Transformers
GGUF
English
mergekit
Merge
Inference Endpoints
imatrix
conversational
Model card Files Community
mradermacher commited on
Commit
c8afeaa
•
1 Parent(s): dea3075

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +9 -0
README.md CHANGED
@@ -32,10 +32,19 @@ more details, including on how to concatenate multi-part files.
32
 
33
  | Link | Type | Size/GB | Notes |
34
  |:-----|:-----|--------:|:------|
 
35
  | [GGUF](https://huggingface.co/mradermacher/Cathallama-70B-i1-GGUF/resolve/main/Cathallama-70B.i1-Q2_K.gguf) | i1-Q2_K | 26.5 | IQ3_XXS probably better |
36
  | [GGUF](https://huggingface.co/mradermacher/Cathallama-70B-i1-GGUF/resolve/main/Cathallama-70B.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 27.6 | lower quality |
 
37
  | [GGUF](https://huggingface.co/mradermacher/Cathallama-70B-i1-GGUF/resolve/main/Cathallama-70B.i1-Q3_K_M.gguf) | i1-Q3_K_M | 34.4 | IQ3_S probably better |
 
 
 
38
  | [GGUF](https://huggingface.co/mradermacher/Cathallama-70B-i1-GGUF/resolve/main/Cathallama-70B.i1-Q4_K_S.gguf) | i1-Q4_K_S | 40.4 | optimal size/speed/quality |
 
 
 
 
39
 
40
  Here is a handy graph by ikawrakow comparing some lower-quality quant
41
  types (lower is better):
 
32
 
33
  | Link | Type | Size/GB | Notes |
34
  |:-----|:-----|--------:|:------|
35
+ | [GGUF](https://huggingface.co/mradermacher/Cathallama-70B-i1-GGUF/resolve/main/Cathallama-70B.i1-IQ2_M.gguf) | i1-IQ2_M | 24.2 | |
36
  | [GGUF](https://huggingface.co/mradermacher/Cathallama-70B-i1-GGUF/resolve/main/Cathallama-70B.i1-Q2_K.gguf) | i1-Q2_K | 26.5 | IQ3_XXS probably better |
37
  | [GGUF](https://huggingface.co/mradermacher/Cathallama-70B-i1-GGUF/resolve/main/Cathallama-70B.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 27.6 | lower quality |
38
+ | [GGUF](https://huggingface.co/mradermacher/Cathallama-70B-i1-GGUF/resolve/main/Cathallama-70B.i1-Q3_K_S.gguf) | i1-Q3_K_S | 31.0 | IQ3_XS probably better |
39
  | [GGUF](https://huggingface.co/mradermacher/Cathallama-70B-i1-GGUF/resolve/main/Cathallama-70B.i1-Q3_K_M.gguf) | i1-Q3_K_M | 34.4 | IQ3_S probably better |
40
+ | [GGUF](https://huggingface.co/mradermacher/Cathallama-70B-i1-GGUF/resolve/main/Cathallama-70B.i1-Q3_K_L.gguf) | i1-Q3_K_L | 37.2 | IQ3_M probably better |
41
+ | [GGUF](https://huggingface.co/mradermacher/Cathallama-70B-i1-GGUF/resolve/main/Cathallama-70B.i1-IQ4_XS.gguf) | i1-IQ4_XS | 38.0 | |
42
+ | [GGUF](https://huggingface.co/mradermacher/Cathallama-70B-i1-GGUF/resolve/main/Cathallama-70B.i1-Q4_0.gguf) | i1-Q4_0 | 40.2 | fast, low quality |
43
  | [GGUF](https://huggingface.co/mradermacher/Cathallama-70B-i1-GGUF/resolve/main/Cathallama-70B.i1-Q4_K_S.gguf) | i1-Q4_K_S | 40.4 | optimal size/speed/quality |
44
+ | [GGUF](https://huggingface.co/mradermacher/Cathallama-70B-i1-GGUF/resolve/main/Cathallama-70B.i1-Q4_K_M.gguf) | i1-Q4_K_M | 42.6 | fast, recommended |
45
+ | [GGUF](https://huggingface.co/mradermacher/Cathallama-70B-i1-GGUF/resolve/main/Cathallama-70B.i1-Q5_K_S.gguf) | i1-Q5_K_S | 48.8 | |
46
+ | [GGUF](https://huggingface.co/mradermacher/Cathallama-70B-i1-GGUF/resolve/main/Cathallama-70B.i1-Q5_K_M.gguf) | i1-Q5_K_M | 50.0 | |
47
+ | [PART 1](https://huggingface.co/mradermacher/Cathallama-70B-i1-GGUF/resolve/main/Cathallama-70B.i1-Q6_K.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Cathallama-70B-i1-GGUF/resolve/main/Cathallama-70B.i1-Q6_K.gguf.part2of2) | i1-Q6_K | 58.0 | practically like static Q6_K |
48
 
49
  Here is a handy graph by ikawrakow comparing some lower-quality quant
50
  types (lower is better):