mradermacher commited on
Commit
960d6ad
1 Parent(s): 3c5c2e3

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +15 -0
README.md CHANGED
@@ -43,12 +43,27 @@ more details, including on how to concatenate multi-part files.
43
 
44
  | Link | Type | Size/GB | Notes |
45
  |:-----|:-----|--------:|:------|
 
 
 
 
 
 
46
  | [GGUF](https://huggingface.co/mradermacher/ChatWaifu_22B_v2.0_preview-i1-GGUF/resolve/main/ChatWaifu_22B_v2.0_preview.i1-Q2_K.gguf) | i1-Q2_K | 8.4 | IQ3_XXS probably better |
47
  | [GGUF](https://huggingface.co/mradermacher/ChatWaifu_22B_v2.0_preview-i1-GGUF/resolve/main/ChatWaifu_22B_v2.0_preview.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 8.7 | lower quality |
 
 
 
48
  | [GGUF](https://huggingface.co/mradermacher/ChatWaifu_22B_v2.0_preview-i1-GGUF/resolve/main/ChatWaifu_22B_v2.0_preview.i1-IQ3_M.gguf) | i1-IQ3_M | 10.2 | |
49
  | [GGUF](https://huggingface.co/mradermacher/ChatWaifu_22B_v2.0_preview-i1-GGUF/resolve/main/ChatWaifu_22B_v2.0_preview.i1-Q3_K_M.gguf) | i1-Q3_K_M | 10.9 | IQ3_S probably better |
 
 
 
50
  | [GGUF](https://huggingface.co/mradermacher/ChatWaifu_22B_v2.0_preview-i1-GGUF/resolve/main/ChatWaifu_22B_v2.0_preview.i1-Q4_K_S.gguf) | i1-Q4_K_S | 12.8 | optimal size/speed/quality |
51
  | [GGUF](https://huggingface.co/mradermacher/ChatWaifu_22B_v2.0_preview-i1-GGUF/resolve/main/ChatWaifu_22B_v2.0_preview.i1-Q4_K_M.gguf) | i1-Q4_K_M | 13.4 | fast, recommended |
 
 
 
52
 
53
  Here is a handy graph by ikawrakow comparing some lower-quality quant
54
  types (lower is better):
 
43
 
44
  | Link | Type | Size/GB | Notes |
45
  |:-----|:-----|--------:|:------|
46
+ | [GGUF](https://huggingface.co/mradermacher/ChatWaifu_22B_v2.0_preview-i1-GGUF/resolve/main/ChatWaifu_22B_v2.0_preview.i1-IQ1_S.gguf) | i1-IQ1_S | 4.9 | for the desperate |
47
+ | [GGUF](https://huggingface.co/mradermacher/ChatWaifu_22B_v2.0_preview-i1-GGUF/resolve/main/ChatWaifu_22B_v2.0_preview.i1-IQ1_M.gguf) | i1-IQ1_M | 5.4 | mostly desperate |
48
+ | [GGUF](https://huggingface.co/mradermacher/ChatWaifu_22B_v2.0_preview-i1-GGUF/resolve/main/ChatWaifu_22B_v2.0_preview.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 6.1 | |
49
+ | [GGUF](https://huggingface.co/mradermacher/ChatWaifu_22B_v2.0_preview-i1-GGUF/resolve/main/ChatWaifu_22B_v2.0_preview.i1-IQ2_XS.gguf) | i1-IQ2_XS | 6.7 | |
50
+ | [GGUF](https://huggingface.co/mradermacher/ChatWaifu_22B_v2.0_preview-i1-GGUF/resolve/main/ChatWaifu_22B_v2.0_preview.i1-IQ2_S.gguf) | i1-IQ2_S | 7.1 | |
51
+ | [GGUF](https://huggingface.co/mradermacher/ChatWaifu_22B_v2.0_preview-i1-GGUF/resolve/main/ChatWaifu_22B_v2.0_preview.i1-IQ2_M.gguf) | i1-IQ2_M | 7.7 | |
52
  | [GGUF](https://huggingface.co/mradermacher/ChatWaifu_22B_v2.0_preview-i1-GGUF/resolve/main/ChatWaifu_22B_v2.0_preview.i1-Q2_K.gguf) | i1-Q2_K | 8.4 | IQ3_XXS probably better |
53
  | [GGUF](https://huggingface.co/mradermacher/ChatWaifu_22B_v2.0_preview-i1-GGUF/resolve/main/ChatWaifu_22B_v2.0_preview.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 8.7 | lower quality |
54
+ | [GGUF](https://huggingface.co/mradermacher/ChatWaifu_22B_v2.0_preview-i1-GGUF/resolve/main/ChatWaifu_22B_v2.0_preview.i1-IQ3_XS.gguf) | i1-IQ3_XS | 9.3 | |
55
+ | [GGUF](https://huggingface.co/mradermacher/ChatWaifu_22B_v2.0_preview-i1-GGUF/resolve/main/ChatWaifu_22B_v2.0_preview.i1-Q3_K_S.gguf) | i1-Q3_K_S | 9.7 | IQ3_XS probably better |
56
+ | [GGUF](https://huggingface.co/mradermacher/ChatWaifu_22B_v2.0_preview-i1-GGUF/resolve/main/ChatWaifu_22B_v2.0_preview.i1-IQ3_S.gguf) | i1-IQ3_S | 9.8 | beats Q3_K* |
57
  | [GGUF](https://huggingface.co/mradermacher/ChatWaifu_22B_v2.0_preview-i1-GGUF/resolve/main/ChatWaifu_22B_v2.0_preview.i1-IQ3_M.gguf) | i1-IQ3_M | 10.2 | |
58
  | [GGUF](https://huggingface.co/mradermacher/ChatWaifu_22B_v2.0_preview-i1-GGUF/resolve/main/ChatWaifu_22B_v2.0_preview.i1-Q3_K_M.gguf) | i1-Q3_K_M | 10.9 | IQ3_S probably better |
59
+ | [GGUF](https://huggingface.co/mradermacher/ChatWaifu_22B_v2.0_preview-i1-GGUF/resolve/main/ChatWaifu_22B_v2.0_preview.i1-Q3_K_L.gguf) | i1-Q3_K_L | 11.8 | IQ3_M probably better |
60
+ | [GGUF](https://huggingface.co/mradermacher/ChatWaifu_22B_v2.0_preview-i1-GGUF/resolve/main/ChatWaifu_22B_v2.0_preview.i1-IQ4_XS.gguf) | i1-IQ4_XS | 12.0 | |
61
+ | [GGUF](https://huggingface.co/mradermacher/ChatWaifu_22B_v2.0_preview-i1-GGUF/resolve/main/ChatWaifu_22B_v2.0_preview.i1-Q4_0.gguf) | i1-Q4_0 | 12.7 | fast, low quality |
62
  | [GGUF](https://huggingface.co/mradermacher/ChatWaifu_22B_v2.0_preview-i1-GGUF/resolve/main/ChatWaifu_22B_v2.0_preview.i1-Q4_K_S.gguf) | i1-Q4_K_S | 12.8 | optimal size/speed/quality |
63
  | [GGUF](https://huggingface.co/mradermacher/ChatWaifu_22B_v2.0_preview-i1-GGUF/resolve/main/ChatWaifu_22B_v2.0_preview.i1-Q4_K_M.gguf) | i1-Q4_K_M | 13.4 | fast, recommended |
64
+ | [GGUF](https://huggingface.co/mradermacher/ChatWaifu_22B_v2.0_preview-i1-GGUF/resolve/main/ChatWaifu_22B_v2.0_preview.i1-Q5_K_S.gguf) | i1-Q5_K_S | 15.4 | |
65
+ | [GGUF](https://huggingface.co/mradermacher/ChatWaifu_22B_v2.0_preview-i1-GGUF/resolve/main/ChatWaifu_22B_v2.0_preview.i1-Q5_K_M.gguf) | i1-Q5_K_M | 15.8 | |
66
+ | [GGUF](https://huggingface.co/mradermacher/ChatWaifu_22B_v2.0_preview-i1-GGUF/resolve/main/ChatWaifu_22B_v2.0_preview.i1-Q6_K.gguf) | i1-Q6_K | 18.4 | practically like static Q6_K |
67
 
68
  Here is a handy graph by ikawrakow comparing some lower-quality quant
69
  types (lower is better):