mradermacher commited on
Commit
1d4cada
·
verified ·
1 Parent(s): 48190f4

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +8 -0
README.md CHANGED
@@ -43,11 +43,19 @@ more details, including on how to concatenate multi-part files.
43
  | Link | Type | Size/GB | Notes |
44
  |:-----|:-----|--------:|:------|
45
  | [GGUF](https://huggingface.co/mradermacher/WEBGEN-Devstral-24B-i1-GGUF/resolve/main/WEBGEN-Devstral-24B.imatrix.gguf) | imatrix | 0.1 | imatrix file (for creating your own qwuants) |
 
 
 
 
46
  | [GGUF](https://huggingface.co/mradermacher/WEBGEN-Devstral-24B-i1-GGUF/resolve/main/WEBGEN-Devstral-24B.i1-Q2_K.gguf) | i1-Q2_K | 9.0 | IQ3_XXS probably better |
 
47
  | [GGUF](https://huggingface.co/mradermacher/WEBGEN-Devstral-24B-i1-GGUF/resolve/main/WEBGEN-Devstral-24B.i1-IQ3_M.gguf) | i1-IQ3_M | 10.8 | |
48
  | [GGUF](https://huggingface.co/mradermacher/WEBGEN-Devstral-24B-i1-GGUF/resolve/main/WEBGEN-Devstral-24B.i1-Q3_K_M.gguf) | i1-Q3_K_M | 11.6 | IQ3_S probably better |
 
 
49
  | [GGUF](https://huggingface.co/mradermacher/WEBGEN-Devstral-24B-i1-GGUF/resolve/main/WEBGEN-Devstral-24B.i1-Q4_K_S.gguf) | i1-Q4_K_S | 13.6 | optimal size/speed/quality |
50
  | [GGUF](https://huggingface.co/mradermacher/WEBGEN-Devstral-24B-i1-GGUF/resolve/main/WEBGEN-Devstral-24B.i1-Q4_K_M.gguf) | i1-Q4_K_M | 14.4 | fast, recommended |
 
51
 
52
  Here is a handy graph by ikawrakow comparing some lower-quality quant
53
  types (lower is better):
 
43
  | Link | Type | Size/GB | Notes |
44
  |:-----|:-----|--------:|:------|
45
  | [GGUF](https://huggingface.co/mradermacher/WEBGEN-Devstral-24B-i1-GGUF/resolve/main/WEBGEN-Devstral-24B.imatrix.gguf) | imatrix | 0.1 | imatrix file (for creating your own qwuants) |
46
+ | [GGUF](https://huggingface.co/mradermacher/WEBGEN-Devstral-24B-i1-GGUF/resolve/main/WEBGEN-Devstral-24B.i1-IQ1_M.gguf) | i1-IQ1_M | 5.9 | mostly desperate |
47
+ | [GGUF](https://huggingface.co/mradermacher/WEBGEN-Devstral-24B-i1-GGUF/resolve/main/WEBGEN-Devstral-24B.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 6.6 | |
48
+ | [GGUF](https://huggingface.co/mradermacher/WEBGEN-Devstral-24B-i1-GGUF/resolve/main/WEBGEN-Devstral-24B.i1-IQ2_M.gguf) | i1-IQ2_M | 8.2 | |
49
+ | [GGUF](https://huggingface.co/mradermacher/WEBGEN-Devstral-24B-i1-GGUF/resolve/main/WEBGEN-Devstral-24B.i1-Q2_K_S.gguf) | i1-Q2_K_S | 8.4 | very low quality |
50
  | [GGUF](https://huggingface.co/mradermacher/WEBGEN-Devstral-24B-i1-GGUF/resolve/main/WEBGEN-Devstral-24B.i1-Q2_K.gguf) | i1-Q2_K | 9.0 | IQ3_XXS probably better |
51
+ | [GGUF](https://huggingface.co/mradermacher/WEBGEN-Devstral-24B-i1-GGUF/resolve/main/WEBGEN-Devstral-24B.i1-Q3_K_S.gguf) | i1-Q3_K_S | 10.5 | IQ3_XS probably better |
52
  | [GGUF](https://huggingface.co/mradermacher/WEBGEN-Devstral-24B-i1-GGUF/resolve/main/WEBGEN-Devstral-24B.i1-IQ3_M.gguf) | i1-IQ3_M | 10.8 | |
53
  | [GGUF](https://huggingface.co/mradermacher/WEBGEN-Devstral-24B-i1-GGUF/resolve/main/WEBGEN-Devstral-24B.i1-Q3_K_M.gguf) | i1-Q3_K_M | 11.6 | IQ3_S probably better |
54
+ | [GGUF](https://huggingface.co/mradermacher/WEBGEN-Devstral-24B-i1-GGUF/resolve/main/WEBGEN-Devstral-24B.i1-Q3_K_L.gguf) | i1-Q3_K_L | 12.5 | IQ3_M probably better |
55
+ | [GGUF](https://huggingface.co/mradermacher/WEBGEN-Devstral-24B-i1-GGUF/resolve/main/WEBGEN-Devstral-24B.i1-IQ4_XS.gguf) | i1-IQ4_XS | 12.9 | |
56
  | [GGUF](https://huggingface.co/mradermacher/WEBGEN-Devstral-24B-i1-GGUF/resolve/main/WEBGEN-Devstral-24B.i1-Q4_K_S.gguf) | i1-Q4_K_S | 13.6 | optimal size/speed/quality |
57
  | [GGUF](https://huggingface.co/mradermacher/WEBGEN-Devstral-24B-i1-GGUF/resolve/main/WEBGEN-Devstral-24B.i1-Q4_K_M.gguf) | i1-Q4_K_M | 14.4 | fast, recommended |
58
+ | [GGUF](https://huggingface.co/mradermacher/WEBGEN-Devstral-24B-i1-GGUF/resolve/main/WEBGEN-Devstral-24B.i1-Q6_K.gguf) | i1-Q6_K | 19.4 | practically like static Q6_K |
59
 
60
  Here is a handy graph by ikawrakow comparing some lower-quality quant
61
  types (lower is better):