Hugging Face's logo

mradermacher
/
GEITje-7B-ultra-sft-GGUF

Transformers
GGUF
Dutch
alignment-handbook
trl
sft
geitje
conversational
Inference Endpoints
Model card Files Community
mradermacher commited on
Commit
f235923
·
verified ·
1 Parent(s): 5520dcf

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +4 -0
README.md CHANGED
@@ -45,10 +45,14 @@ more details, including on how to concatenate multi-part files.
45
  | [GGUF](https://huggingface.co/mradermacher/GEITje-7B-ultra-sft-GGUF/resolve/main/GEITje-7B-ultra-sft.Q3_K_S.gguf) | Q3_K_S | 3.3 | |
46
  | [GGUF](https://huggingface.co/mradermacher/GEITje-7B-ultra-sft-GGUF/resolve/main/GEITje-7B-ultra-sft.Q3_K_M.gguf) | Q3_K_M | 3.6 | lower quality |
47
  | [GGUF](https://huggingface.co/mradermacher/GEITje-7B-ultra-sft-GGUF/resolve/main/GEITje-7B-ultra-sft.Q3_K_L.gguf) | Q3_K_L | 3.9 | |
 
48
  | [GGUF](https://huggingface.co/mradermacher/GEITje-7B-ultra-sft-GGUF/resolve/main/GEITje-7B-ultra-sft.Q4_K_S.gguf) | Q4_K_S | 4.2 | fast, recommended |
49
  | [GGUF](https://huggingface.co/mradermacher/GEITje-7B-ultra-sft-GGUF/resolve/main/GEITje-7B-ultra-sft.Q4_K_M.gguf) | Q4_K_M | 4.5 | fast, recommended |
 
 
50
  | [GGUF](https://huggingface.co/mradermacher/GEITje-7B-ultra-sft-GGUF/resolve/main/GEITje-7B-ultra-sft.Q6_K.gguf) | Q6_K | 6.0 | very good quality |
51
  | [GGUF](https://huggingface.co/mradermacher/GEITje-7B-ultra-sft-GGUF/resolve/main/GEITje-7B-ultra-sft.Q8_0.gguf) | Q8_0 | 7.8 | fast, best quality |
 
52
 
53
  Here is a handy graph by ikawrakow comparing some lower-quality quant
54
  types (lower is better):
 
45
  | [GGUF](https://huggingface.co/mradermacher/GEITje-7B-ultra-sft-GGUF/resolve/main/GEITje-7B-ultra-sft.Q3_K_S.gguf) | Q3_K_S | 3.3 | |
46
  | [GGUF](https://huggingface.co/mradermacher/GEITje-7B-ultra-sft-GGUF/resolve/main/GEITje-7B-ultra-sft.Q3_K_M.gguf) | Q3_K_M | 3.6 | lower quality |
47
  | [GGUF](https://huggingface.co/mradermacher/GEITje-7B-ultra-sft-GGUF/resolve/main/GEITje-7B-ultra-sft.Q3_K_L.gguf) | Q3_K_L | 3.9 | |
48
+ | [GGUF](https://huggingface.co/mradermacher/GEITje-7B-ultra-sft-GGUF/resolve/main/GEITje-7B-ultra-sft.IQ4_XS.gguf) | IQ4_XS | 4.0 | |
49
  | [GGUF](https://huggingface.co/mradermacher/GEITje-7B-ultra-sft-GGUF/resolve/main/GEITje-7B-ultra-sft.Q4_K_S.gguf) | Q4_K_S | 4.2 | fast, recommended |
50
  | [GGUF](https://huggingface.co/mradermacher/GEITje-7B-ultra-sft-GGUF/resolve/main/GEITje-7B-ultra-sft.Q4_K_M.gguf) | Q4_K_M | 4.5 | fast, recommended |
51
+ | [GGUF](https://huggingface.co/mradermacher/GEITje-7B-ultra-sft-GGUF/resolve/main/GEITje-7B-ultra-sft.Q5_K_S.gguf) | Q5_K_S | 5.1 | |
52
+ | [GGUF](https://huggingface.co/mradermacher/GEITje-7B-ultra-sft-GGUF/resolve/main/GEITje-7B-ultra-sft.Q5_K_M.gguf) | Q5_K_M | 5.2 | |
53
  | [GGUF](https://huggingface.co/mradermacher/GEITje-7B-ultra-sft-GGUF/resolve/main/GEITje-7B-ultra-sft.Q6_K.gguf) | Q6_K | 6.0 | very good quality |
54
  | [GGUF](https://huggingface.co/mradermacher/GEITje-7B-ultra-sft-GGUF/resolve/main/GEITje-7B-ultra-sft.Q8_0.gguf) | Q8_0 | 7.8 | fast, best quality |
55
+ | [GGUF](https://huggingface.co/mradermacher/GEITje-7B-ultra-sft-GGUF/resolve/main/GEITje-7B-ultra-sft.f16.gguf) | f16 | 14.6 | 16 bpw, overkill |
56
 
57
  Here is a handy graph by ikawrakow comparing some lower-quality quant
58
  types (lower is better):