mradermacher commited on
Commit
1a887d4
·
verified ·
1 Parent(s): 6efccba

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +6 -0
README.md CHANGED
@@ -35,7 +35,13 @@ more details, including on how to concatenate multi-part files.
35
  | [GGUF](https://huggingface.co/mradermacher/mistral-nemo-narwhal-12B-GGUF/resolve/main/mistral-nemo-narwhal-12B.Q3_K_S.gguf) | Q3_K_S | 5.6 | |
36
  | [GGUF](https://huggingface.co/mradermacher/mistral-nemo-narwhal-12B-GGUF/resolve/main/mistral-nemo-narwhal-12B.Q3_K_M.gguf) | Q3_K_M | 6.2 | lower quality |
37
  | [GGUF](https://huggingface.co/mradermacher/mistral-nemo-narwhal-12B-GGUF/resolve/main/mistral-nemo-narwhal-12B.Q3_K_L.gguf) | Q3_K_L | 6.7 | |
 
38
  | [GGUF](https://huggingface.co/mradermacher/mistral-nemo-narwhal-12B-GGUF/resolve/main/mistral-nemo-narwhal-12B.Q4_K_S.gguf) | Q4_K_S | 7.2 | fast, recommended |
 
 
 
 
 
39
 
40
  Here is a handy graph by ikawrakow comparing some lower-quality quant
41
  types (lower is better):
 
35
  | [GGUF](https://huggingface.co/mradermacher/mistral-nemo-narwhal-12B-GGUF/resolve/main/mistral-nemo-narwhal-12B.Q3_K_S.gguf) | Q3_K_S | 5.6 | |
36
  | [GGUF](https://huggingface.co/mradermacher/mistral-nemo-narwhal-12B-GGUF/resolve/main/mistral-nemo-narwhal-12B.Q3_K_M.gguf) | Q3_K_M | 6.2 | lower quality |
37
  | [GGUF](https://huggingface.co/mradermacher/mistral-nemo-narwhal-12B-GGUF/resolve/main/mistral-nemo-narwhal-12B.Q3_K_L.gguf) | Q3_K_L | 6.7 | |
38
+ | [GGUF](https://huggingface.co/mradermacher/mistral-nemo-narwhal-12B-GGUF/resolve/main/mistral-nemo-narwhal-12B.IQ4_XS.gguf) | IQ4_XS | 6.9 | |
39
  | [GGUF](https://huggingface.co/mradermacher/mistral-nemo-narwhal-12B-GGUF/resolve/main/mistral-nemo-narwhal-12B.Q4_K_S.gguf) | Q4_K_S | 7.2 | fast, recommended |
40
+ | [GGUF](https://huggingface.co/mradermacher/mistral-nemo-narwhal-12B-GGUF/resolve/main/mistral-nemo-narwhal-12B.Q4_K_M.gguf) | Q4_K_M | 7.6 | fast, recommended |
41
+ | [GGUF](https://huggingface.co/mradermacher/mistral-nemo-narwhal-12B-GGUF/resolve/main/mistral-nemo-narwhal-12B.Q5_K_S.gguf) | Q5_K_S | 8.6 | |
42
+ | [GGUF](https://huggingface.co/mradermacher/mistral-nemo-narwhal-12B-GGUF/resolve/main/mistral-nemo-narwhal-12B.Q5_K_M.gguf) | Q5_K_M | 8.8 | |
43
+ | [GGUF](https://huggingface.co/mradermacher/mistral-nemo-narwhal-12B-GGUF/resolve/main/mistral-nemo-narwhal-12B.Q6_K.gguf) | Q6_K | 10.2 | very good quality |
44
+ | [GGUF](https://huggingface.co/mradermacher/mistral-nemo-narwhal-12B-GGUF/resolve/main/mistral-nemo-narwhal-12B.Q8_0.gguf) | Q8_0 | 13.1 | fast, best quality |
45
 
46
  Here is a handy graph by ikawrakow comparing some lower-quality quant
47
  types (lower is better):