Hugging Face's logo

mradermacher
/
Q2.5-32B-Slush-i1-GGUF

Transformers
GGUF
English
Not-For-All-Audiences
mergekit
Inference Endpoints
imatrix
conversational
Model card Files Community
mradermacher commited on
Commit
5db9760
•
1 Parent(s): cffc430

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +9 -0
README.md CHANGED
@@ -38,17 +38,26 @@ more details, including on how to concatenate multi-part files.
38
 
39
  | Link | Type | Size/GB | Notes |
40
  |:-----|:-----|--------:|:------|
 
41
  | [GGUF](https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF/resolve/main/Q2.5-32B-Slush.i1-IQ1_M.gguf) | i1-IQ1_M | 8.0 | mostly desperate |
42
  | [GGUF](https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF/resolve/main/Q2.5-32B-Slush.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 9.1 | |
 
 
43
  | [GGUF](https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF/resolve/main/Q2.5-32B-Slush.i1-IQ2_M.gguf) | i1-IQ2_M | 11.4 | |
44
  | [GGUF](https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF/resolve/main/Q2.5-32B-Slush.i1-Q2_K.gguf) | i1-Q2_K | 12.4 | IQ3_XXS probably better |
45
  | [GGUF](https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF/resolve/main/Q2.5-32B-Slush.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 12.9 | lower quality |
 
46
  | [GGUF](https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF/resolve/main/Q2.5-32B-Slush.i1-Q3_K_S.gguf) | i1-Q3_K_S | 14.5 | IQ3_XS probably better |
 
47
  | [GGUF](https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF/resolve/main/Q2.5-32B-Slush.i1-IQ3_M.gguf) | i1-IQ3_M | 14.9 | |
48
  | [GGUF](https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF/resolve/main/Q2.5-32B-Slush.i1-Q3_K_M.gguf) | i1-Q3_K_M | 16.0 | IQ3_S probably better |
 
49
  | [GGUF](https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF/resolve/main/Q2.5-32B-Slush.i1-IQ4_XS.gguf) | i1-IQ4_XS | 17.8 | |
 
50
  | [GGUF](https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF/resolve/main/Q2.5-32B-Slush.i1-Q4_K_S.gguf) | i1-Q4_K_S | 18.9 | optimal size/speed/quality |
51
  | [GGUF](https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF/resolve/main/Q2.5-32B-Slush.i1-Q4_K_M.gguf) | i1-Q4_K_M | 19.9 | fast, recommended |
 
 
52
  | [GGUF](https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF/resolve/main/Q2.5-32B-Slush.i1-Q6_K.gguf) | i1-Q6_K | 27.0 | practically like static Q6_K |
53
 
54
  Here is a handy graph by ikawrakow comparing some lower-quality quant
 
38
 
39
  | Link | Type | Size/GB | Notes |
40
  |:-----|:-----|--------:|:------|
41
+ | [GGUF](https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF/resolve/main/Q2.5-32B-Slush.i1-IQ1_S.gguf) | i1-IQ1_S | 7.4 | for the desperate |
42
  | [GGUF](https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF/resolve/main/Q2.5-32B-Slush.i1-IQ1_M.gguf) | i1-IQ1_M | 8.0 | mostly desperate |
43
  | [GGUF](https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF/resolve/main/Q2.5-32B-Slush.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 9.1 | |
44
+ | [GGUF](https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF/resolve/main/Q2.5-32B-Slush.i1-IQ2_XS.gguf) | i1-IQ2_XS | 10.1 | |
45
+ | [GGUF](https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF/resolve/main/Q2.5-32B-Slush.i1-IQ2_S.gguf) | i1-IQ2_S | 10.5 | |
46
  | [GGUF](https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF/resolve/main/Q2.5-32B-Slush.i1-IQ2_M.gguf) | i1-IQ2_M | 11.4 | |
47
  | [GGUF](https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF/resolve/main/Q2.5-32B-Slush.i1-Q2_K.gguf) | i1-Q2_K | 12.4 | IQ3_XXS probably better |
48
  | [GGUF](https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF/resolve/main/Q2.5-32B-Slush.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 12.9 | lower quality |
49
+ | [GGUF](https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF/resolve/main/Q2.5-32B-Slush.i1-IQ3_XS.gguf) | i1-IQ3_XS | 13.8 | |
50
  | [GGUF](https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF/resolve/main/Q2.5-32B-Slush.i1-Q3_K_S.gguf) | i1-Q3_K_S | 14.5 | IQ3_XS probably better |
51
+ | [GGUF](https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF/resolve/main/Q2.5-32B-Slush.i1-IQ3_S.gguf) | i1-IQ3_S | 14.5 | beats Q3_K* |
52
  | [GGUF](https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF/resolve/main/Q2.5-32B-Slush.i1-IQ3_M.gguf) | i1-IQ3_M | 14.9 | |
53
  | [GGUF](https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF/resolve/main/Q2.5-32B-Slush.i1-Q3_K_M.gguf) | i1-Q3_K_M | 16.0 | IQ3_S probably better |
54
+ | [GGUF](https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF/resolve/main/Q2.5-32B-Slush.i1-Q3_K_L.gguf) | i1-Q3_K_L | 17.3 | IQ3_M probably better |
55
  | [GGUF](https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF/resolve/main/Q2.5-32B-Slush.i1-IQ4_XS.gguf) | i1-IQ4_XS | 17.8 | |
56
+ | [GGUF](https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF/resolve/main/Q2.5-32B-Slush.i1-Q4_0.gguf) | i1-Q4_0 | 18.8 | fast, low quality |
57
  | [GGUF](https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF/resolve/main/Q2.5-32B-Slush.i1-Q4_K_S.gguf) | i1-Q4_K_S | 18.9 | optimal size/speed/quality |
58
  | [GGUF](https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF/resolve/main/Q2.5-32B-Slush.i1-Q4_K_M.gguf) | i1-Q4_K_M | 19.9 | fast, recommended |
59
+ | [GGUF](https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF/resolve/main/Q2.5-32B-Slush.i1-Q5_K_S.gguf) | i1-Q5_K_S | 22.7 | |
60
+ | [GGUF](https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF/resolve/main/Q2.5-32B-Slush.i1-Q5_K_M.gguf) | i1-Q5_K_M | 23.4 | |
61
  | [GGUF](https://huggingface.co/mradermacher/Q2.5-32B-Slush-i1-GGUF/resolve/main/Q2.5-32B-Slush.i1-Q6_K.gguf) | i1-Q6_K | 27.0 | practically like static Q6_K |
62
 
63
  Here is a handy graph by ikawrakow comparing some lower-quality quant