mradermacher commited on
Commit
5147013
·
verified ·
1 Parent(s): ddb8391

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +15 -0
README.md CHANGED
@@ -39,12 +39,27 @@ more details, including on how to concatenate multi-part files.
39
 
40
  | Link | Type | Size/GB | Notes |
41
  |:-----|:-----|--------:|:------|
 
 
 
 
 
 
42
  | [GGUF](https://huggingface.co/mradermacher/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3-i1-GGUF/resolve/main/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3.i1-Q2_K.gguf) | i1-Q2_K | 8.9 | IQ3_XXS probably better |
43
  | [GGUF](https://huggingface.co/mradermacher/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3-i1-GGUF/resolve/main/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 9.4 | lower quality |
 
 
 
44
  | [GGUF](https://huggingface.co/mradermacher/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3-i1-GGUF/resolve/main/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3.i1-IQ3_M.gguf) | i1-IQ3_M | 10.8 | |
45
  | [GGUF](https://huggingface.co/mradermacher/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3-i1-GGUF/resolve/main/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3.i1-Q3_K_M.gguf) | i1-Q3_K_M | 11.7 | IQ3_S probably better |
 
 
 
46
  | [GGUF](https://huggingface.co/mradermacher/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3-i1-GGUF/resolve/main/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3.i1-Q4_K_S.gguf) | i1-Q4_K_S | 13.8 | optimal size/speed/quality |
47
  | [GGUF](https://huggingface.co/mradermacher/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3-i1-GGUF/resolve/main/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3.i1-Q4_K_M.gguf) | i1-Q4_K_M | 14.7 | fast, recommended |
 
 
 
48
 
49
  Here is a handy graph by ikawrakow comparing some lower-quality quant
50
  types (lower is better):
 
39
 
40
  | Link | Type | Size/GB | Notes |
41
  |:-----|:-----|--------:|:------|
42
+ | [GGUF](https://huggingface.co/mradermacher/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3-i1-GGUF/resolve/main/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3.i1-IQ1_S.gguf) | i1-IQ1_S | 5.1 | for the desperate |
43
+ | [GGUF](https://huggingface.co/mradermacher/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3-i1-GGUF/resolve/main/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3.i1-IQ1_M.gguf) | i1-IQ1_M | 5.6 | mostly desperate |
44
+ | [GGUF](https://huggingface.co/mradermacher/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3-i1-GGUF/resolve/main/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 6.5 | |
45
+ | [GGUF](https://huggingface.co/mradermacher/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3-i1-GGUF/resolve/main/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3.i1-IQ2_XS.gguf) | i1-IQ2_XS | 7.2 | |
46
+ | [GGUF](https://huggingface.co/mradermacher/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3-i1-GGUF/resolve/main/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3.i1-IQ2_S.gguf) | i1-IQ2_S | 7.4 | |
47
+ | [GGUF](https://huggingface.co/mradermacher/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3-i1-GGUF/resolve/main/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3.i1-IQ2_M.gguf) | i1-IQ2_M | 8.1 | |
48
  | [GGUF](https://huggingface.co/mradermacher/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3-i1-GGUF/resolve/main/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3.i1-Q2_K.gguf) | i1-Q2_K | 8.9 | IQ3_XXS probably better |
49
  | [GGUF](https://huggingface.co/mradermacher/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3-i1-GGUF/resolve/main/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 9.4 | lower quality |
50
+ | [GGUF](https://huggingface.co/mradermacher/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3-i1-GGUF/resolve/main/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3.i1-IQ3_XS.gguf) | i1-IQ3_XS | 10.0 | |
51
+ | [GGUF](https://huggingface.co/mradermacher/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3-i1-GGUF/resolve/main/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3.i1-Q3_K_S.gguf) | i1-Q3_K_S | 10.5 | IQ3_XS probably better |
52
+ | [GGUF](https://huggingface.co/mradermacher/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3-i1-GGUF/resolve/main/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3.i1-IQ3_S.gguf) | i1-IQ3_S | 10.6 | beats Q3_K* |
53
  | [GGUF](https://huggingface.co/mradermacher/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3-i1-GGUF/resolve/main/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3.i1-IQ3_M.gguf) | i1-IQ3_M | 10.8 | |
54
  | [GGUF](https://huggingface.co/mradermacher/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3-i1-GGUF/resolve/main/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3.i1-Q3_K_M.gguf) | i1-Q3_K_M | 11.7 | IQ3_S probably better |
55
+ | [GGUF](https://huggingface.co/mradermacher/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3-i1-GGUF/resolve/main/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3.i1-Q3_K_L.gguf) | i1-Q3_K_L | 12.6 | IQ3_M probably better |
56
+ | [GGUF](https://huggingface.co/mradermacher/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3-i1-GGUF/resolve/main/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3.i1-IQ4_XS.gguf) | i1-IQ4_XS | 13.0 | |
57
+ | [GGUF](https://huggingface.co/mradermacher/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3-i1-GGUF/resolve/main/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3.i1-Q4_0.gguf) | i1-Q4_0 | 13.8 | fast, low quality |
58
  | [GGUF](https://huggingface.co/mradermacher/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3-i1-GGUF/resolve/main/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3.i1-Q4_K_S.gguf) | i1-Q4_K_S | 13.8 | optimal size/speed/quality |
59
  | [GGUF](https://huggingface.co/mradermacher/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3-i1-GGUF/resolve/main/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3.i1-Q4_K_M.gguf) | i1-Q4_K_M | 14.7 | fast, recommended |
60
+ | [GGUF](https://huggingface.co/mradermacher/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3-i1-GGUF/resolve/main/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3.i1-Q5_K_S.gguf) | i1-Q5_K_S | 16.7 | |
61
+ | [GGUF](https://huggingface.co/mradermacher/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3-i1-GGUF/resolve/main/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3.i1-Q5_K_M.gguf) | i1-Q5_K_M | 17.2 | |
62
+ | [GGUF](https://huggingface.co/mradermacher/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3-i1-GGUF/resolve/main/MixtureofMerges-MoE-4x7b-v10-mixtralv0.3.i1-Q6_K.gguf) | i1-Q6_K | 19.9 | practically like static Q6_K |
63
 
64
  Here is a handy graph by ikawrakow comparing some lower-quality quant
65
  types (lower is better):