mradermacher commited on
Commit
84f391a
·
verified ·
1 Parent(s): 5ca722c

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +5 -0
README.md CHANGED
@@ -51,9 +51,14 @@ more details, including on how to concatenate multi-part files.
51
  | [GGUF](https://huggingface.co/mradermacher/Sailor2-8B-GGUF/resolve/main/Sailor2-8B.Q3_K_S.gguf) | Q3_K_S | 4.0 | |
52
  | [GGUF](https://huggingface.co/mradermacher/Sailor2-8B-GGUF/resolve/main/Sailor2-8B.Q3_K_M.gguf) | Q3_K_M | 4.4 | lower quality |
53
  | [GGUF](https://huggingface.co/mradermacher/Sailor2-8B-GGUF/resolve/main/Sailor2-8B.Q3_K_L.gguf) | Q3_K_L | 4.7 | |
 
54
  | [GGUF](https://huggingface.co/mradermacher/Sailor2-8B-GGUF/resolve/main/Sailor2-8B.Q4_K_S.gguf) | Q4_K_S | 5.1 | fast, recommended |
 
 
 
55
  | [GGUF](https://huggingface.co/mradermacher/Sailor2-8B-GGUF/resolve/main/Sailor2-8B.Q6_K.gguf) | Q6_K | 7.1 | very good quality |
56
  | [GGUF](https://huggingface.co/mradermacher/Sailor2-8B-GGUF/resolve/main/Sailor2-8B.Q8_0.gguf) | Q8_0 | 9.2 | fast, best quality |
 
57
 
58
  Here is a handy graph by ikawrakow comparing some lower-quality quant
59
  types (lower is better):
 
51
  | [GGUF](https://huggingface.co/mradermacher/Sailor2-8B-GGUF/resolve/main/Sailor2-8B.Q3_K_S.gguf) | Q3_K_S | 4.0 | |
52
  | [GGUF](https://huggingface.co/mradermacher/Sailor2-8B-GGUF/resolve/main/Sailor2-8B.Q3_K_M.gguf) | Q3_K_M | 4.4 | lower quality |
53
  | [GGUF](https://huggingface.co/mradermacher/Sailor2-8B-GGUF/resolve/main/Sailor2-8B.Q3_K_L.gguf) | Q3_K_L | 4.7 | |
54
+ | [GGUF](https://huggingface.co/mradermacher/Sailor2-8B-GGUF/resolve/main/Sailor2-8B.IQ4_XS.gguf) | IQ4_XS | 4.9 | |
55
  | [GGUF](https://huggingface.co/mradermacher/Sailor2-8B-GGUF/resolve/main/Sailor2-8B.Q4_K_S.gguf) | Q4_K_S | 5.1 | fast, recommended |
56
+ | [GGUF](https://huggingface.co/mradermacher/Sailor2-8B-GGUF/resolve/main/Sailor2-8B.Q4_K_M.gguf) | Q4_K_M | 5.3 | fast, recommended |
57
+ | [GGUF](https://huggingface.co/mradermacher/Sailor2-8B-GGUF/resolve/main/Sailor2-8B.Q5_K_S.gguf) | Q5_K_S | 6.1 | |
58
+ | [GGUF](https://huggingface.co/mradermacher/Sailor2-8B-GGUF/resolve/main/Sailor2-8B.Q5_K_M.gguf) | Q5_K_M | 6.2 | |
59
  | [GGUF](https://huggingface.co/mradermacher/Sailor2-8B-GGUF/resolve/main/Sailor2-8B.Q6_K.gguf) | Q6_K | 7.1 | very good quality |
60
  | [GGUF](https://huggingface.co/mradermacher/Sailor2-8B-GGUF/resolve/main/Sailor2-8B.Q8_0.gguf) | Q8_0 | 9.2 | fast, best quality |
61
+ | [GGUF](https://huggingface.co/mradermacher/Sailor2-8B-GGUF/resolve/main/Sailor2-8B.f16.gguf) | f16 | 17.2 | 16 bpw, overkill |
62
 
63
  Here is a handy graph by ikawrakow comparing some lower-quality quant
64
  types (lower is better):