mradermacher commited on
Commit
4446a45
1 Parent(s): 932f64d

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +14 -0
README.md CHANGED
@@ -44,13 +44,27 @@ more details, including on how to concatenate multi-part files.
44
 
45
  | Link | Type | Size/GB | Notes |
46
  |:-----|:-----|--------:|:------|
 
 
 
 
 
47
  | [GGUF](https://huggingface.co/mradermacher/airoboros-34b-3.2-i1-GGUF/resolve/main/airoboros-34b-3.2.i1-IQ2_M.gguf) | i1-IQ2_M | 11.9 | |
48
  | [GGUF](https://huggingface.co/mradermacher/airoboros-34b-3.2-i1-GGUF/resolve/main/airoboros-34b-3.2.i1-Q2_K.gguf) | i1-Q2_K | 12.9 | IQ3_XXS probably better |
49
  | [GGUF](https://huggingface.co/mradermacher/airoboros-34b-3.2-i1-GGUF/resolve/main/airoboros-34b-3.2.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 13.4 | lower quality |
 
 
 
50
  | [GGUF](https://huggingface.co/mradermacher/airoboros-34b-3.2-i1-GGUF/resolve/main/airoboros-34b-3.2.i1-IQ3_M.gguf) | i1-IQ3_M | 15.7 | |
51
  | [GGUF](https://huggingface.co/mradermacher/airoboros-34b-3.2-i1-GGUF/resolve/main/airoboros-34b-3.2.i1-Q3_K_M.gguf) | i1-Q3_K_M | 16.8 | IQ3_S probably better |
 
 
 
52
  | [GGUF](https://huggingface.co/mradermacher/airoboros-34b-3.2-i1-GGUF/resolve/main/airoboros-34b-3.2.i1-Q4_K_S.gguf) | i1-Q4_K_S | 19.7 | optimal size/speed/quality |
53
  | [GGUF](https://huggingface.co/mradermacher/airoboros-34b-3.2-i1-GGUF/resolve/main/airoboros-34b-3.2.i1-Q4_K_M.gguf) | i1-Q4_K_M | 20.8 | fast, recommended |
 
 
 
54
 
55
  Here is a handy graph by ikawrakow comparing some lower-quality quant
56
  types (lower is better):
 
44
 
45
  | Link | Type | Size/GB | Notes |
46
  |:-----|:-----|--------:|:------|
47
+ | [GGUF](https://huggingface.co/mradermacher/airoboros-34b-3.2-i1-GGUF/resolve/main/airoboros-34b-3.2.i1-IQ1_S.gguf) | i1-IQ1_S | 7.6 | for the desperate |
48
+ | [GGUF](https://huggingface.co/mradermacher/airoboros-34b-3.2-i1-GGUF/resolve/main/airoboros-34b-3.2.i1-IQ1_M.gguf) | i1-IQ1_M | 8.3 | mostly desperate |
49
+ | [GGUF](https://huggingface.co/mradermacher/airoboros-34b-3.2-i1-GGUF/resolve/main/airoboros-34b-3.2.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 9.4 | |
50
+ | [GGUF](https://huggingface.co/mradermacher/airoboros-34b-3.2-i1-GGUF/resolve/main/airoboros-34b-3.2.i1-IQ2_XS.gguf) | i1-IQ2_XS | 10.4 | |
51
+ | [GGUF](https://huggingface.co/mradermacher/airoboros-34b-3.2-i1-GGUF/resolve/main/airoboros-34b-3.2.i1-IQ2_S.gguf) | i1-IQ2_S | 11.0 | |
52
  | [GGUF](https://huggingface.co/mradermacher/airoboros-34b-3.2-i1-GGUF/resolve/main/airoboros-34b-3.2.i1-IQ2_M.gguf) | i1-IQ2_M | 11.9 | |
53
  | [GGUF](https://huggingface.co/mradermacher/airoboros-34b-3.2-i1-GGUF/resolve/main/airoboros-34b-3.2.i1-Q2_K.gguf) | i1-Q2_K | 12.9 | IQ3_XXS probably better |
54
  | [GGUF](https://huggingface.co/mradermacher/airoboros-34b-3.2-i1-GGUF/resolve/main/airoboros-34b-3.2.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 13.4 | lower quality |
55
+ | [GGUF](https://huggingface.co/mradermacher/airoboros-34b-3.2-i1-GGUF/resolve/main/airoboros-34b-3.2.i1-IQ3_XS.gguf) | i1-IQ3_XS | 14.3 | |
56
+ | [GGUF](https://huggingface.co/mradermacher/airoboros-34b-3.2-i1-GGUF/resolve/main/airoboros-34b-3.2.i1-Q3_K_S.gguf) | i1-Q3_K_S | 15.1 | IQ3_XS probably better |
57
+ | [GGUF](https://huggingface.co/mradermacher/airoboros-34b-3.2-i1-GGUF/resolve/main/airoboros-34b-3.2.i1-IQ3_S.gguf) | i1-IQ3_S | 15.1 | beats Q3_K* |
58
  | [GGUF](https://huggingface.co/mradermacher/airoboros-34b-3.2-i1-GGUF/resolve/main/airoboros-34b-3.2.i1-IQ3_M.gguf) | i1-IQ3_M | 15.7 | |
59
  | [GGUF](https://huggingface.co/mradermacher/airoboros-34b-3.2-i1-GGUF/resolve/main/airoboros-34b-3.2.i1-Q3_K_M.gguf) | i1-Q3_K_M | 16.8 | IQ3_S probably better |
60
+ | [GGUF](https://huggingface.co/mradermacher/airoboros-34b-3.2-i1-GGUF/resolve/main/airoboros-34b-3.2.i1-Q3_K_L.gguf) | i1-Q3_K_L | 18.2 | IQ3_M probably better |
61
+ | [GGUF](https://huggingface.co/mradermacher/airoboros-34b-3.2-i1-GGUF/resolve/main/airoboros-34b-3.2.i1-IQ4_XS.gguf) | i1-IQ4_XS | 18.6 | |
62
+ | [GGUF](https://huggingface.co/mradermacher/airoboros-34b-3.2-i1-GGUF/resolve/main/airoboros-34b-3.2.i1-Q4_0.gguf) | i1-Q4_0 | 19.6 | fast, low quality |
63
  | [GGUF](https://huggingface.co/mradermacher/airoboros-34b-3.2-i1-GGUF/resolve/main/airoboros-34b-3.2.i1-Q4_K_S.gguf) | i1-Q4_K_S | 19.7 | optimal size/speed/quality |
64
  | [GGUF](https://huggingface.co/mradermacher/airoboros-34b-3.2-i1-GGUF/resolve/main/airoboros-34b-3.2.i1-Q4_K_M.gguf) | i1-Q4_K_M | 20.8 | fast, recommended |
65
+ | [GGUF](https://huggingface.co/mradermacher/airoboros-34b-3.2-i1-GGUF/resolve/main/airoboros-34b-3.2.i1-Q5_K_S.gguf) | i1-Q5_K_S | 23.8 | |
66
+ | [GGUF](https://huggingface.co/mradermacher/airoboros-34b-3.2-i1-GGUF/resolve/main/airoboros-34b-3.2.i1-Q5_K_M.gguf) | i1-Q5_K_M | 24.4 | |
67
+ | [GGUF](https://huggingface.co/mradermacher/airoboros-34b-3.2-i1-GGUF/resolve/main/airoboros-34b-3.2.i1-Q6_K.gguf) | i1-Q6_K | 28.3 | practically like static Q6_K |
68
 
69
  Here is a handy graph by ikawrakow comparing some lower-quality quant
70
  types (lower is better):