Transformers
GGUF
imatrix
mradermacher commited on
Commit
9905612
·
verified ·
1 Parent(s): 7545e9a

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +18 -18
README.md CHANGED
@@ -39,24 +39,24 @@ more details, including on how to concatenate multi-part files.
39
 
40
  | Link | Type | Size/GB | Notes |
41
  |:-----|:-----|--------:|:------|
42
- | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 9.3 | |
43
- | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/.i1-IQ2_XS.gguf) | i1-IQ2_XS | 10.2 | |
44
- | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/.i1-IQ2_M.gguf) | i1-IQ2_M | 11.7 | |
45
- | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/.i1-Q2_K.gguf) | i1-Q2_K | 12.7 | IQ3_XXS probably better |
46
- | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 13.1 | lower quality |
47
- | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/.i1-IQ3_XS.gguf) | i1-IQ3_XS | 14.0 | |
48
- | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/.i1-Q3_K_S.gguf) | i1-Q3_K_S | 14.7 | IQ3_XS probably better |
49
- | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/.i1-IQ3_S.gguf) | i1-IQ3_S | 14.8 | beats Q3_K* |
50
- | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/.i1-IQ3_M.gguf) | i1-IQ3_M | 15.3 | |
51
- | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/.i1-Q3_K_M.gguf) | i1-Q3_K_M | 16.3 | IQ3_S probably better |
52
- | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/.i1-Q3_K_L.gguf) | i1-Q3_K_L | 17.7 | IQ3_M probably better |
53
- | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/.i1-IQ4_XS.gguf) | i1-IQ4_XS | 18.0 | |
54
- | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/.i1-Q4_0.gguf) | i1-Q4_0 | 19.0 | fast, low quality |
55
- | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/.i1-Q4_K_S.gguf) | i1-Q4_K_S | 19.1 | optimal size/speed/quality |
56
- | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/.i1-Q4_K_M.gguf) | i1-Q4_K_M | 20.1 | fast, recommended |
57
- | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/.i1-Q5_K_S.gguf) | i1-Q5_K_S | 23.0 | |
58
- | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/.i1-Q5_K_M.gguf) | i1-Q5_K_M | 23.6 | |
59
- | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/.i1-Q6_K.gguf) | i1-Q6_K | 27.3 | practically like static Q6_K |
60
 
61
  Here is a handy graph by ikawrakow comparing some lower-quality quant
62
  types (lower is better):
 
39
 
40
  | Link | Type | Size/GB | Notes |
41
  |:-----|:-----|--------:|:------|
42
+ | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/Viking-33B.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 9.3 | |
43
+ | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/Viking-33B.i1-IQ2_XS.gguf) | i1-IQ2_XS | 10.2 | |
44
+ | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/Viking-33B.i1-IQ2_M.gguf) | i1-IQ2_M | 11.7 | |
45
+ | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/Viking-33B.i1-Q2_K.gguf) | i1-Q2_K | 12.7 | IQ3_XXS probably better |
46
+ | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/Viking-33B.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 13.1 | lower quality |
47
+ | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/Viking-33B.i1-IQ3_XS.gguf) | i1-IQ3_XS | 14.0 | |
48
+ | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/Viking-33B.i1-Q3_K_S.gguf) | i1-Q3_K_S | 14.7 | IQ3_XS probably better |
49
+ | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/Viking-33B.i1-IQ3_S.gguf) | i1-IQ3_S | 14.8 | beats Q3_K* |
50
+ | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/Viking-33B.i1-IQ3_M.gguf) | i1-IQ3_M | 15.3 | |
51
+ | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/Viking-33B.i1-Q3_K_M.gguf) | i1-Q3_K_M | 16.3 | IQ3_S probably better |
52
+ | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/Viking-33B.i1-Q3_K_L.gguf) | i1-Q3_K_L | 17.7 | IQ3_M probably better |
53
+ | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/Viking-33B.i1-IQ4_XS.gguf) | i1-IQ4_XS | 18.0 | |
54
+ | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/Viking-33B.i1-Q4_0.gguf) | i1-Q4_0 | 19.0 | fast, low quality |
55
+ | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/Viking-33B.i1-Q4_K_S.gguf) | i1-Q4_K_S | 19.1 | optimal size/speed/quality |
56
+ | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/Viking-33B.i1-Q4_K_M.gguf) | i1-Q4_K_M | 20.1 | fast, recommended |
57
+ | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/Viking-33B.i1-Q5_K_S.gguf) | i1-Q5_K_S | 23.0 | |
58
+ | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/Viking-33B.i1-Q5_K_M.gguf) | i1-Q5_K_M | 23.6 | |
59
+ | [GGUF](https://huggingface.co/mradermacher/Viking-33B-i1-GGUF/resolve/main/Viking-33B.i1-Q6_K.gguf) | i1-Q6_K | 27.3 | practically like static Q6_K |
60
 
61
  Here is a handy graph by ikawrakow comparing some lower-quality quant
62
  types (lower is better):