Transformers
GGUF
Arabic
English
imatrix
conversational
mradermacher commited on
Commit
09c3af7
·
verified ·
1 Parent(s): e3449b1

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +16 -0
README.md CHANGED
@@ -34,12 +34,28 @@ more details, including on how to concatenate multi-part files.
34
 
35
  | Link | Type | Size/GB | Notes |
36
  |:-----|:-----|--------:|:------|
 
 
 
 
 
 
 
37
  | [GGUF](https://huggingface.co/mradermacher/SambaLingo-Arabic-Chat-70B-i1-GGUF/resolve/main/SambaLingo-Arabic-Chat-70B.i1-Q2_K.gguf) | i1-Q2_K | 25.8 | IQ3_XXS probably better |
38
  | [GGUF](https://huggingface.co/mradermacher/SambaLingo-Arabic-Chat-70B-i1-GGUF/resolve/main/SambaLingo-Arabic-Chat-70B.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 26.9 | lower quality |
 
 
 
39
  | [GGUF](https://huggingface.co/mradermacher/SambaLingo-Arabic-Chat-70B-i1-GGUF/resolve/main/SambaLingo-Arabic-Chat-70B.i1-IQ3_M.gguf) | i1-IQ3_M | 31.3 | |
40
  | [GGUF](https://huggingface.co/mradermacher/SambaLingo-Arabic-Chat-70B-i1-GGUF/resolve/main/SambaLingo-Arabic-Chat-70B.i1-Q3_K_M.gguf) | i1-Q3_K_M | 33.6 | IQ3_S probably better |
 
 
 
41
  | [GGUF](https://huggingface.co/mradermacher/SambaLingo-Arabic-Chat-70B-i1-GGUF/resolve/main/SambaLingo-Arabic-Chat-70B.i1-Q4_K_S.gguf) | i1-Q4_K_S | 39.6 | optimal size/speed/quality |
42
  | [GGUF](https://huggingface.co/mradermacher/SambaLingo-Arabic-Chat-70B-i1-GGUF/resolve/main/SambaLingo-Arabic-Chat-70B.i1-Q4_K_M.gguf) | i1-Q4_K_M | 41.8 | fast, recommended |
 
 
 
43
 
44
  Here is a handy graph by ikawrakow comparing some lower-quality quant
45
  types (lower is better):
 
34
 
35
  | Link | Type | Size/GB | Notes |
36
  |:-----|:-----|--------:|:------|
37
+ | [GGUF](https://huggingface.co/mradermacher/SambaLingo-Arabic-Chat-70B-i1-GGUF/resolve/main/SambaLingo-Arabic-Chat-70B.i1-IQ1_S.gguf) | i1-IQ1_S | 14.8 | for the desperate |
38
+ | [GGUF](https://huggingface.co/mradermacher/SambaLingo-Arabic-Chat-70B-i1-GGUF/resolve/main/SambaLingo-Arabic-Chat-70B.i1-IQ1_M.gguf) | i1-IQ1_M | 16.3 | mostly desperate |
39
+ | [GGUF](https://huggingface.co/mradermacher/SambaLingo-Arabic-Chat-70B-i1-GGUF/resolve/main/SambaLingo-Arabic-Chat-70B.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 18.6 | |
40
+ | [GGUF](https://huggingface.co/mradermacher/SambaLingo-Arabic-Chat-70B-i1-GGUF/resolve/main/SambaLingo-Arabic-Chat-70B.i1-IQ2_XS.gguf) | i1-IQ2_XS | 20.6 | |
41
+ | [GGUF](https://huggingface.co/mradermacher/SambaLingo-Arabic-Chat-70B-i1-GGUF/resolve/main/SambaLingo-Arabic-Chat-70B.i1-IQ2_S.gguf) | i1-IQ2_S | 21.7 | |
42
+ | [GGUF](https://huggingface.co/mradermacher/SambaLingo-Arabic-Chat-70B-i1-GGUF/resolve/main/SambaLingo-Arabic-Chat-70B.i1-IQ2_M.gguf) | i1-IQ2_M | 23.6 | |
43
+ | [GGUF](https://huggingface.co/mradermacher/SambaLingo-Arabic-Chat-70B-i1-GGUF/resolve/main/SambaLingo-Arabic-Chat-70B.i1-Q2_K_S.gguf) | i1-Q2_K_S | 23.9 | very low quality |
44
  | [GGUF](https://huggingface.co/mradermacher/SambaLingo-Arabic-Chat-70B-i1-GGUF/resolve/main/SambaLingo-Arabic-Chat-70B.i1-Q2_K.gguf) | i1-Q2_K | 25.8 | IQ3_XXS probably better |
45
  | [GGUF](https://huggingface.co/mradermacher/SambaLingo-Arabic-Chat-70B-i1-GGUF/resolve/main/SambaLingo-Arabic-Chat-70B.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 26.9 | lower quality |
46
+ | [GGUF](https://huggingface.co/mradermacher/SambaLingo-Arabic-Chat-70B-i1-GGUF/resolve/main/SambaLingo-Arabic-Chat-70B.i1-IQ3_XS.gguf) | i1-IQ3_XS | 28.7 | |
47
+ | [GGUF](https://huggingface.co/mradermacher/SambaLingo-Arabic-Chat-70B-i1-GGUF/resolve/main/SambaLingo-Arabic-Chat-70B.i1-IQ3_S.gguf) | i1-IQ3_S | 30.3 | beats Q3_K* |
48
+ | [GGUF](https://huggingface.co/mradermacher/SambaLingo-Arabic-Chat-70B-i1-GGUF/resolve/main/SambaLingo-Arabic-Chat-70B.i1-Q3_K_S.gguf) | i1-Q3_K_S | 30.3 | IQ3_XS probably better |
49
  | [GGUF](https://huggingface.co/mradermacher/SambaLingo-Arabic-Chat-70B-i1-GGUF/resolve/main/SambaLingo-Arabic-Chat-70B.i1-IQ3_M.gguf) | i1-IQ3_M | 31.3 | |
50
  | [GGUF](https://huggingface.co/mradermacher/SambaLingo-Arabic-Chat-70B-i1-GGUF/resolve/main/SambaLingo-Arabic-Chat-70B.i1-Q3_K_M.gguf) | i1-Q3_K_M | 33.6 | IQ3_S probably better |
51
+ | [GGUF](https://huggingface.co/mradermacher/SambaLingo-Arabic-Chat-70B-i1-GGUF/resolve/main/SambaLingo-Arabic-Chat-70B.i1-Q3_K_L.gguf) | i1-Q3_K_L | 36.5 | IQ3_M probably better |
52
+ | [GGUF](https://huggingface.co/mradermacher/SambaLingo-Arabic-Chat-70B-i1-GGUF/resolve/main/SambaLingo-Arabic-Chat-70B.i1-IQ4_XS.gguf) | i1-IQ4_XS | 37.2 | |
53
+ | [GGUF](https://huggingface.co/mradermacher/SambaLingo-Arabic-Chat-70B-i1-GGUF/resolve/main/SambaLingo-Arabic-Chat-70B.i1-Q4_0.gguf) | i1-Q4_0 | 39.4 | fast, low quality |
54
  | [GGUF](https://huggingface.co/mradermacher/SambaLingo-Arabic-Chat-70B-i1-GGUF/resolve/main/SambaLingo-Arabic-Chat-70B.i1-Q4_K_S.gguf) | i1-Q4_K_S | 39.6 | optimal size/speed/quality |
55
  | [GGUF](https://huggingface.co/mradermacher/SambaLingo-Arabic-Chat-70B-i1-GGUF/resolve/main/SambaLingo-Arabic-Chat-70B.i1-Q4_K_M.gguf) | i1-Q4_K_M | 41.8 | fast, recommended |
56
+ | [GGUF](https://huggingface.co/mradermacher/SambaLingo-Arabic-Chat-70B-i1-GGUF/resolve/main/SambaLingo-Arabic-Chat-70B.i1-Q5_K_S.gguf) | i1-Q5_K_S | 47.9 | |
57
+ | [GGUF](https://huggingface.co/mradermacher/SambaLingo-Arabic-Chat-70B-i1-GGUF/resolve/main/SambaLingo-Arabic-Chat-70B.i1-Q5_K_M.gguf) | i1-Q5_K_M | 49.2 | |
58
+ | [PART 1](https://huggingface.co/mradermacher/SambaLingo-Arabic-Chat-70B-i1-GGUF/resolve/main/SambaLingo-Arabic-Chat-70B.i1-Q6_K.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/SambaLingo-Arabic-Chat-70B-i1-GGUF/resolve/main/SambaLingo-Arabic-Chat-70B.i1-Q6_K.gguf.part2of2) | i1-Q6_K | 57.0 | practically like static Q6_K |
59
 
60
  Here is a handy graph by ikawrakow comparing some lower-quality quant
61
  types (lower is better):