mradermacher commited on
Commit
876d28c
1 Parent(s): 89a6b5d

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +9 -0
README.md CHANGED
@@ -39,20 +39,29 @@ more details, including on how to concatenate multi-part files.
39
 
40
  | Link | Type | Size/GB | Notes |
41
  |:-----|:-----|--------:|:------|
 
42
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-IQ1_M.gguf) | i1-IQ1_M | 1.7 | mostly desperate |
43
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 1.9 | |
44
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-IQ2_XS.gguf) | i1-IQ2_XS | 2.1 | |
 
45
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-IQ2_M.gguf) | i1-IQ2_M | 2.4 | |
46
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-Q2_K.gguf) | i1-Q2_K | 2.7 | IQ3_XXS probably better |
47
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 2.7 | lower quality |
 
 
48
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-Q3_K_S.gguf) | i1-Q3_K_S | 3.0 | IQ3_XS probably better |
49
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-IQ3_M.gguf) | i1-IQ3_M | 3.4 | |
50
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-Q3_K_M.gguf) | i1-Q3_K_M | 3.6 | IQ3_S probably better |
51
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-IQ4_XS.gguf) | i1-IQ4_XS | 3.7 | |
 
 
 
 
52
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-Q4_K_S.gguf) | i1-Q4_K_S | 3.9 | optimal size/speed/quality |
53
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-Q3_K_L.gguf) | i1-Q3_K_L | 3.9 | IQ3_M probably better |
54
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-Q4_K_M.gguf) | i1-Q4_K_M | 4.4 | fast, recommended |
55
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-Q5_K_S.gguf) | i1-Q5_K_S | 4.7 | |
 
56
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-Q6_K.gguf) | i1-Q6_K | 5.6 | practically like static Q6_K |
57
 
58
  Here is a handy graph by ikawrakow comparing some lower-quality quant
 
39
 
40
  | Link | Type | Size/GB | Notes |
41
  |:-----|:-----|--------:|:------|
42
+ | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-IQ1_S.gguf) | i1-IQ1_S | 1.6 | for the desperate |
43
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-IQ1_M.gguf) | i1-IQ1_M | 1.7 | mostly desperate |
44
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 1.9 | |
45
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-IQ2_XS.gguf) | i1-IQ2_XS | 2.1 | |
46
+ | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-IQ2_S.gguf) | i1-IQ2_S | 2.2 | |
47
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-IQ2_M.gguf) | i1-IQ2_M | 2.4 | |
48
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-Q2_K.gguf) | i1-Q2_K | 2.7 | IQ3_XXS probably better |
49
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 2.7 | lower quality |
50
+ | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-IQ3_XS.gguf) | i1-IQ3_XS | 3.0 | |
51
+ | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-IQ3_S.gguf) | i1-IQ3_S | 3.0 | beats Q3_K* |
52
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-Q3_K_S.gguf) | i1-Q3_K_S | 3.0 | IQ3_XS probably better |
53
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-IQ3_M.gguf) | i1-IQ3_M | 3.4 | |
54
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-Q3_K_M.gguf) | i1-Q3_K_M | 3.6 | IQ3_S probably better |
55
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-IQ4_XS.gguf) | i1-IQ4_XS | 3.7 | |
56
+ | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-Q4_0_4_4.gguf) | i1-Q4_0_4_4 | 3.9 | fast on arm, low quality |
57
+ | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-Q4_0_4_8.gguf) | i1-Q4_0_4_8 | 3.9 | fast on arm+i8mm, low quality |
58
+ | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-Q4_0_8_8.gguf) | i1-Q4_0_8_8 | 3.9 | fast on arm+sve, low quality |
59
+ | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-Q4_0.gguf) | i1-Q4_0 | 3.9 | fast, low quality |
60
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-Q4_K_S.gguf) | i1-Q4_K_S | 3.9 | optimal size/speed/quality |
61
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-Q3_K_L.gguf) | i1-Q3_K_L | 3.9 | IQ3_M probably better |
62
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-Q4_K_M.gguf) | i1-Q4_K_M | 4.4 | fast, recommended |
63
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-Q5_K_S.gguf) | i1-Q5_K_S | 4.7 | |
64
+ | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-Q5_K_M.gguf) | i1-Q5_K_M | 5.1 | |
65
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-chat-i1-GGUF/resolve/main/mpt-7b-chat.i1-Q6_K.gguf) | i1-Q6_K | 5.6 | practically like static Q6_K |
66
 
67
  Here is a handy graph by ikawrakow comparing some lower-quality quant