Hugging Face's logo

mradermacher
/
mpt-7b-i1-GGUF

Transformers
GGUF
English
Composer
MosaicML
llm-foundry
StreamingDatasets
Inference Endpoints
imatrix
Model card Files Community
mradermacher commited on
Commit
e546d83
•
1 Parent(s): 23c9b6e

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +7 -0
README.md CHANGED
@@ -40,19 +40,26 @@ more details, including on how to concatenate multi-part files.
40
 
41
  | Link | Type | Size/GB | Notes |
42
  |:-----|:-----|--------:|:------|
 
43
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-IQ1_M.gguf) | i1-IQ1_M | 1.7 | mostly desperate |
44
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 1.9 | |
45
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-IQ2_XS.gguf) | i1-IQ2_XS | 2.1 | |
 
46
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-IQ2_M.gguf) | i1-IQ2_M | 2.4 | |
47
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-Q2_K.gguf) | i1-Q2_K | 2.7 | IQ3_XXS probably better |
48
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 2.7 | lower quality |
 
 
49
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-Q3_K_S.gguf) | i1-Q3_K_S | 3.0 | IQ3_XS probably better |
50
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-IQ3_M.gguf) | i1-IQ3_M | 3.4 | |
51
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-Q3_K_M.gguf) | i1-Q3_K_M | 3.6 | IQ3_S probably better |
52
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-IQ4_XS.gguf) | i1-IQ4_XS | 3.7 | |
 
53
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-Q4_K_S.gguf) | i1-Q4_K_S | 3.9 | optimal size/speed/quality |
54
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-Q3_K_L.gguf) | i1-Q3_K_L | 3.9 | IQ3_M probably better |
55
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-Q4_K_M.gguf) | i1-Q4_K_M | 4.4 | fast, recommended |
 
 
56
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.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
 
40
 
41
  | Link | Type | Size/GB | Notes |
42
  |:-----|:-----|--------:|:------|
43
+ | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-IQ1_S.gguf) | i1-IQ1_S | 1.6 | for the desperate |
44
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-IQ1_M.gguf) | i1-IQ1_M | 1.7 | mostly desperate |
45
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 1.9 | |
46
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-IQ2_XS.gguf) | i1-IQ2_XS | 2.1 | |
47
+ | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-IQ2_S.gguf) | i1-IQ2_S | 2.2 | |
48
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-IQ2_M.gguf) | i1-IQ2_M | 2.4 | |
49
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-Q2_K.gguf) | i1-Q2_K | 2.7 | IQ3_XXS probably better |
50
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 2.7 | lower quality |
51
+ | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-IQ3_XS.gguf) | i1-IQ3_XS | 3.0 | |
52
+ | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-IQ3_S.gguf) | i1-IQ3_S | 3.0 | beats Q3_K* |
53
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-Q3_K_S.gguf) | i1-Q3_K_S | 3.0 | IQ3_XS probably better |
54
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-IQ3_M.gguf) | i1-IQ3_M | 3.4 | |
55
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-Q3_K_M.gguf) | i1-Q3_K_M | 3.6 | IQ3_S probably better |
56
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-IQ4_XS.gguf) | i1-IQ4_XS | 3.7 | |
57
+ | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-Q4_0.gguf) | i1-Q4_0 | 3.9 | fast, low quality |
58
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-Q4_K_S.gguf) | i1-Q4_K_S | 3.9 | optimal size/speed/quality |
59
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-Q3_K_L.gguf) | i1-Q3_K_L | 3.9 | IQ3_M probably better |
60
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-Q4_K_M.gguf) | i1-Q4_K_M | 4.4 | fast, recommended |
61
+ | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-Q5_K_S.gguf) | i1-Q5_K_S | 4.7 | |
62
+ | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-Q5_K_M.gguf) | i1-Q5_K_M | 5.1 | |
63
  | [GGUF](https://huggingface.co/mradermacher/mpt-7b-i1-GGUF/resolve/main/mpt-7b.i1-Q6_K.gguf) | i1-Q6_K | 5.6 | practically like static Q6_K |
64
 
65
  Here is a handy graph by ikawrakow comparing some lower-quality quant