Hugging Face's logo

mradermacher
/
ArrowNeo-AME-3x4B-v0.1-MoE-GGUF

Transformers
GGUF
Japanese
English
conversational
Model card Files Community
mradermacher commited on
Commit
692c64a
·
verified ·
1 Parent(s): badc59f

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +12 -1
README.md CHANGED
@@ -19,7 +19,7 @@ quantized_by: mradermacher
19
  static quants of https://huggingface.co/DataPilot/ArrowNeo-AME-3x4B-v0.1-MoE
20
 
21
  <!-- provided-files -->
22
- weighted/imatrix quants seem not to be available (by me) at this time. If they do not show up a week or so after the static ones, I have probably not planned for them. Feel free to request them by opening a Community Discussion.
23
  ## Usage
24
 
25
  If you are unsure how to use GGUF files, refer to one of [TheBloke's
@@ -33,6 +33,17 @@ more details, including on how to concatenate multi-part files.
33
  | Link | Type | Size/GB | Notes |
34
  |:-----|:-----|--------:|:------|
35
  | [GGUF](https://huggingface.co/mradermacher/ArrowNeo-AME-3x4B-v0.1-MoE-GGUF/resolve/main/ArrowNeo-AME-3x4B-v0.1-MoE.Q2_K.gguf) | Q2_K | 3.0 | |
 
 
 
 
 
 
 
 
 
 
 
36
 
37
  Here is a handy graph by ikawrakow comparing some lower-quality quant
38
  types (lower is better):
 
19
  static quants of https://huggingface.co/DataPilot/ArrowNeo-AME-3x4B-v0.1-MoE
20
 
21
  <!-- provided-files -->
22
+ weighted/imatrix quants are available at https://huggingface.co/mradermacher/ArrowNeo-AME-3x4B-v0.1-MoE-i1-GGUF
23
  ## Usage
24
 
25
  If you are unsure how to use GGUF files, refer to one of [TheBloke's
 
33
  | Link | Type | Size/GB | Notes |
34
  |:-----|:-----|--------:|:------|
35
  | [GGUF](https://huggingface.co/mradermacher/ArrowNeo-AME-3x4B-v0.1-MoE-GGUF/resolve/main/ArrowNeo-AME-3x4B-v0.1-MoE.Q2_K.gguf) | Q2_K | 3.0 | |
36
+ | [GGUF](https://huggingface.co/mradermacher/ArrowNeo-AME-3x4B-v0.1-MoE-GGUF/resolve/main/ArrowNeo-AME-3x4B-v0.1-MoE.Q3_K_S.gguf) | Q3_K_S | 3.5 | |
37
+ | [GGUF](https://huggingface.co/mradermacher/ArrowNeo-AME-3x4B-v0.1-MoE-GGUF/resolve/main/ArrowNeo-AME-3x4B-v0.1-MoE.Q3_K_M.gguf) | Q3_K_M | 3.9 | lower quality |
38
+ | [GGUF](https://huggingface.co/mradermacher/ArrowNeo-AME-3x4B-v0.1-MoE-GGUF/resolve/main/ArrowNeo-AME-3x4B-v0.1-MoE.Q3_K_L.gguf) | Q3_K_L | 4.2 | |
39
+ | [GGUF](https://huggingface.co/mradermacher/ArrowNeo-AME-3x4B-v0.1-MoE-GGUF/resolve/main/ArrowNeo-AME-3x4B-v0.1-MoE.IQ4_XS.gguf) | IQ4_XS | 4.3 | |
40
+ | [GGUF](https://huggingface.co/mradermacher/ArrowNeo-AME-3x4B-v0.1-MoE-GGUF/resolve/main/ArrowNeo-AME-3x4B-v0.1-MoE.Q4_K_S.gguf) | Q4_K_S | 4.6 | fast, recommended |
41
+ | [GGUF](https://huggingface.co/mradermacher/ArrowNeo-AME-3x4B-v0.1-MoE-GGUF/resolve/main/ArrowNeo-AME-3x4B-v0.1-MoE.Q4_K_M.gguf) | Q4_K_M | 4.8 | fast, recommended |
42
+ | [GGUF](https://huggingface.co/mradermacher/ArrowNeo-AME-3x4B-v0.1-MoE-GGUF/resolve/main/ArrowNeo-AME-3x4B-v0.1-MoE.Q5_K_S.gguf) | Q5_K_S | 5.5 | |
43
+ | [GGUF](https://huggingface.co/mradermacher/ArrowNeo-AME-3x4B-v0.1-MoE-GGUF/resolve/main/ArrowNeo-AME-3x4B-v0.1-MoE.Q5_K_M.gguf) | Q5_K_M | 5.6 | |
44
+ | [GGUF](https://huggingface.co/mradermacher/ArrowNeo-AME-3x4B-v0.1-MoE-GGUF/resolve/main/ArrowNeo-AME-3x4B-v0.1-MoE.Q6_K.gguf) | Q6_K | 6.5 | very good quality |
45
+ | [GGUF](https://huggingface.co/mradermacher/ArrowNeo-AME-3x4B-v0.1-MoE-GGUF/resolve/main/ArrowNeo-AME-3x4B-v0.1-MoE.Q8_0.gguf) | Q8_0 | 8.3 | fast, best quality |
46
+ | [GGUF](https://huggingface.co/mradermacher/ArrowNeo-AME-3x4B-v0.1-MoE-GGUF/resolve/main/ArrowNeo-AME-3x4B-v0.1-MoE.f16.gguf) | f16 | 15.6 | 16 bpw, overkill |
47
 
48
  Here is a handy graph by ikawrakow comparing some lower-quality quant
49
  types (lower is better):