Hugging Face's logo

mradermacher
/
speed-synthesis-8b-senior-i1-GGUF

Transformers
GGUF
English
Inference Endpoints
imatrix
Model card Files Community
mradermacher commited on
Commit
55d56f2
•
1 Parent(s): cc5257e

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +17 -0
README.md CHANGED
@@ -31,13 +31,30 @@ more details, including on how to concatenate multi-part files.
31
 
32
  | Link | Type | Size/GB | Notes |
33
  |:-----|:-----|--------:|:------|
 
 
 
 
 
34
  | [GGUF](https://huggingface.co/mradermacher/speed-synthesis-8b-senior-i1-GGUF/resolve/main/speed-synthesis-8b-senior.i1-IQ2_M.gguf) | i1-IQ2_M | 3.0 | |
35
  | [GGUF](https://huggingface.co/mradermacher/speed-synthesis-8b-senior-i1-GGUF/resolve/main/speed-synthesis-8b-senior.i1-Q2_K.gguf) | i1-Q2_K | 3.3 | IQ3_XXS probably better |
36
  | [GGUF](https://huggingface.co/mradermacher/speed-synthesis-8b-senior-i1-GGUF/resolve/main/speed-synthesis-8b-senior.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 3.4 | lower quality |
 
 
 
37
  | [GGUF](https://huggingface.co/mradermacher/speed-synthesis-8b-senior-i1-GGUF/resolve/main/speed-synthesis-8b-senior.i1-IQ3_M.gguf) | i1-IQ3_M | 3.9 | |
38
  | [GGUF](https://huggingface.co/mradermacher/speed-synthesis-8b-senior-i1-GGUF/resolve/main/speed-synthesis-8b-senior.i1-Q3_K_M.gguf) | i1-Q3_K_M | 4.1 | IQ3_S probably better |
 
 
 
 
 
 
39
  | [GGUF](https://huggingface.co/mradermacher/speed-synthesis-8b-senior-i1-GGUF/resolve/main/speed-synthesis-8b-senior.i1-Q4_K_S.gguf) | i1-Q4_K_S | 4.8 | optimal size/speed/quality |
40
  | [GGUF](https://huggingface.co/mradermacher/speed-synthesis-8b-senior-i1-GGUF/resolve/main/speed-synthesis-8b-senior.i1-Q4_K_M.gguf) | i1-Q4_K_M | 5.0 | fast, recommended |
 
 
 
41
 
42
  Here is a handy graph by ikawrakow comparing some lower-quality quant
43
  types (lower is better):
 
31
 
32
  | Link | Type | Size/GB | Notes |
33
  |:-----|:-----|--------:|:------|
34
+ | [GGUF](https://huggingface.co/mradermacher/speed-synthesis-8b-senior-i1-GGUF/resolve/main/speed-synthesis-8b-senior.i1-IQ1_S.gguf) | i1-IQ1_S | 2.1 | for the desperate |
35
+ | [GGUF](https://huggingface.co/mradermacher/speed-synthesis-8b-senior-i1-GGUF/resolve/main/speed-synthesis-8b-senior.i1-IQ1_M.gguf) | i1-IQ1_M | 2.3 | mostly desperate |
36
+ | [GGUF](https://huggingface.co/mradermacher/speed-synthesis-8b-senior-i1-GGUF/resolve/main/speed-synthesis-8b-senior.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 2.5 | |
37
+ | [GGUF](https://huggingface.co/mradermacher/speed-synthesis-8b-senior-i1-GGUF/resolve/main/speed-synthesis-8b-senior.i1-IQ2_XS.gguf) | i1-IQ2_XS | 2.7 | |
38
+ | [GGUF](https://huggingface.co/mradermacher/speed-synthesis-8b-senior-i1-GGUF/resolve/main/speed-synthesis-8b-senior.i1-IQ2_S.gguf) | i1-IQ2_S | 2.9 | |
39
  | [GGUF](https://huggingface.co/mradermacher/speed-synthesis-8b-senior-i1-GGUF/resolve/main/speed-synthesis-8b-senior.i1-IQ2_M.gguf) | i1-IQ2_M | 3.0 | |
40
  | [GGUF](https://huggingface.co/mradermacher/speed-synthesis-8b-senior-i1-GGUF/resolve/main/speed-synthesis-8b-senior.i1-Q2_K.gguf) | i1-Q2_K | 3.3 | IQ3_XXS probably better |
41
  | [GGUF](https://huggingface.co/mradermacher/speed-synthesis-8b-senior-i1-GGUF/resolve/main/speed-synthesis-8b-senior.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 3.4 | lower quality |
42
+ | [GGUF](https://huggingface.co/mradermacher/speed-synthesis-8b-senior-i1-GGUF/resolve/main/speed-synthesis-8b-senior.i1-IQ3_XS.gguf) | i1-IQ3_XS | 3.6 | |
43
+ | [GGUF](https://huggingface.co/mradermacher/speed-synthesis-8b-senior-i1-GGUF/resolve/main/speed-synthesis-8b-senior.i1-Q3_K_S.gguf) | i1-Q3_K_S | 3.8 | IQ3_XS probably better |
44
+ | [GGUF](https://huggingface.co/mradermacher/speed-synthesis-8b-senior-i1-GGUF/resolve/main/speed-synthesis-8b-senior.i1-IQ3_S.gguf) | i1-IQ3_S | 3.8 | beats Q3_K* |
45
  | [GGUF](https://huggingface.co/mradermacher/speed-synthesis-8b-senior-i1-GGUF/resolve/main/speed-synthesis-8b-senior.i1-IQ3_M.gguf) | i1-IQ3_M | 3.9 | |
46
  | [GGUF](https://huggingface.co/mradermacher/speed-synthesis-8b-senior-i1-GGUF/resolve/main/speed-synthesis-8b-senior.i1-Q3_K_M.gguf) | i1-Q3_K_M | 4.1 | IQ3_S probably better |
47
+ | [GGUF](https://huggingface.co/mradermacher/speed-synthesis-8b-senior-i1-GGUF/resolve/main/speed-synthesis-8b-senior.i1-Q3_K_L.gguf) | i1-Q3_K_L | 4.4 | IQ3_M probably better |
48
+ | [GGUF](https://huggingface.co/mradermacher/speed-synthesis-8b-senior-i1-GGUF/resolve/main/speed-synthesis-8b-senior.i1-IQ4_XS.gguf) | i1-IQ4_XS | 4.5 | |
49
+ | [GGUF](https://huggingface.co/mradermacher/speed-synthesis-8b-senior-i1-GGUF/resolve/main/speed-synthesis-8b-senior.i1-Q4_0_4_4.gguf) | i1-Q4_0_4_4 | 4.8 | fast on arm, low quality |
50
+ | [GGUF](https://huggingface.co/mradermacher/speed-synthesis-8b-senior-i1-GGUF/resolve/main/speed-synthesis-8b-senior.i1-Q4_0_4_8.gguf) | i1-Q4_0_4_8 | 4.8 | fast on arm+i8mm, low quality |
51
+ | [GGUF](https://huggingface.co/mradermacher/speed-synthesis-8b-senior-i1-GGUF/resolve/main/speed-synthesis-8b-senior.i1-Q4_0_8_8.gguf) | i1-Q4_0_8_8 | 4.8 | fast on arm+sve, low quality |
52
+ | [GGUF](https://huggingface.co/mradermacher/speed-synthesis-8b-senior-i1-GGUF/resolve/main/speed-synthesis-8b-senior.i1-Q4_0.gguf) | i1-Q4_0 | 4.8 | fast, low quality |
53
  | [GGUF](https://huggingface.co/mradermacher/speed-synthesis-8b-senior-i1-GGUF/resolve/main/speed-synthesis-8b-senior.i1-Q4_K_S.gguf) | i1-Q4_K_S | 4.8 | optimal size/speed/quality |
54
  | [GGUF](https://huggingface.co/mradermacher/speed-synthesis-8b-senior-i1-GGUF/resolve/main/speed-synthesis-8b-senior.i1-Q4_K_M.gguf) | i1-Q4_K_M | 5.0 | fast, recommended |
55
+ | [GGUF](https://huggingface.co/mradermacher/speed-synthesis-8b-senior-i1-GGUF/resolve/main/speed-synthesis-8b-senior.i1-Q5_K_S.gguf) | i1-Q5_K_S | 5.7 | |
56
+ | [GGUF](https://huggingface.co/mradermacher/speed-synthesis-8b-senior-i1-GGUF/resolve/main/speed-synthesis-8b-senior.i1-Q5_K_M.gguf) | i1-Q5_K_M | 5.8 | |
57
+ | [GGUF](https://huggingface.co/mradermacher/speed-synthesis-8b-senior-i1-GGUF/resolve/main/speed-synthesis-8b-senior.i1-Q6_K.gguf) | i1-Q6_K | 6.7 | practically like static Q6_K |
58
 
59
  Here is a handy graph by ikawrakow comparing some lower-quality quant
60
  types (lower is better):