Hugging Face's logo

mradermacher
/
Misal-7B-base-v0.1-GGUF

Transformers
GGUF
Marathi
marathi
Inference Endpoints
Model card Files Community
mradermacher commited on
Commit
e050830
·
verified ·
1 Parent(s): ea9288d

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +4 -0
README.md CHANGED
@@ -39,10 +39,14 @@ more details, including on how to concatenate multi-part files.
39
  | [GGUF](https://huggingface.co/mradermacher/Misal-7B-base-v0.1-GGUF/resolve/main/Misal-7B-base-v0.1.Q3_K_S.gguf) | Q3_K_S | 3.1 | |
40
  | [GGUF](https://huggingface.co/mradermacher/Misal-7B-base-v0.1-GGUF/resolve/main/Misal-7B-base-v0.1.Q3_K_M.gguf) | Q3_K_M | 3.5 | lower quality |
41
  | [GGUF](https://huggingface.co/mradermacher/Misal-7B-base-v0.1-GGUF/resolve/main/Misal-7B-base-v0.1.Q3_K_L.gguf) | Q3_K_L | 3.8 | |
 
42
  | [GGUF](https://huggingface.co/mradermacher/Misal-7B-base-v0.1-GGUF/resolve/main/Misal-7B-base-v0.1.Q4_K_S.gguf) | Q4_K_S | 4.0 | fast, recommended |
43
  | [GGUF](https://huggingface.co/mradermacher/Misal-7B-base-v0.1-GGUF/resolve/main/Misal-7B-base-v0.1.Q4_K_M.gguf) | Q4_K_M | 4.3 | fast, recommended |
 
 
44
  | [GGUF](https://huggingface.co/mradermacher/Misal-7B-base-v0.1-GGUF/resolve/main/Misal-7B-base-v0.1.Q6_K.gguf) | Q6_K | 5.7 | very good quality |
45
  | [GGUF](https://huggingface.co/mradermacher/Misal-7B-base-v0.1-GGUF/resolve/main/Misal-7B-base-v0.1.Q8_0.gguf) | Q8_0 | 7.4 | fast, best quality |
 
46
 
47
  Here is a handy graph by ikawrakow comparing some lower-quality quant
48
  types (lower is better):
 
39
  | [GGUF](https://huggingface.co/mradermacher/Misal-7B-base-v0.1-GGUF/resolve/main/Misal-7B-base-v0.1.Q3_K_S.gguf) | Q3_K_S | 3.1 | |
40
  | [GGUF](https://huggingface.co/mradermacher/Misal-7B-base-v0.1-GGUF/resolve/main/Misal-7B-base-v0.1.Q3_K_M.gguf) | Q3_K_M | 3.5 | lower quality |
41
  | [GGUF](https://huggingface.co/mradermacher/Misal-7B-base-v0.1-GGUF/resolve/main/Misal-7B-base-v0.1.Q3_K_L.gguf) | Q3_K_L | 3.8 | |
42
+ | [GGUF](https://huggingface.co/mradermacher/Misal-7B-base-v0.1-GGUF/resolve/main/Misal-7B-base-v0.1.IQ4_XS.gguf) | IQ4_XS | 3.8 | |
43
  | [GGUF](https://huggingface.co/mradermacher/Misal-7B-base-v0.1-GGUF/resolve/main/Misal-7B-base-v0.1.Q4_K_S.gguf) | Q4_K_S | 4.0 | fast, recommended |
44
  | [GGUF](https://huggingface.co/mradermacher/Misal-7B-base-v0.1-GGUF/resolve/main/Misal-7B-base-v0.1.Q4_K_M.gguf) | Q4_K_M | 4.3 | fast, recommended |
45
+ | [GGUF](https://huggingface.co/mradermacher/Misal-7B-base-v0.1-GGUF/resolve/main/Misal-7B-base-v0.1.Q5_K_S.gguf) | Q5_K_S | 4.8 | |
46
+ | [GGUF](https://huggingface.co/mradermacher/Misal-7B-base-v0.1-GGUF/resolve/main/Misal-7B-base-v0.1.Q5_K_M.gguf) | Q5_K_M | 5.0 | |
47
  | [GGUF](https://huggingface.co/mradermacher/Misal-7B-base-v0.1-GGUF/resolve/main/Misal-7B-base-v0.1.Q6_K.gguf) | Q6_K | 5.7 | very good quality |
48
  | [GGUF](https://huggingface.co/mradermacher/Misal-7B-base-v0.1-GGUF/resolve/main/Misal-7B-base-v0.1.Q8_0.gguf) | Q8_0 | 7.4 | fast, best quality |
49
+ | [GGUF](https://huggingface.co/mradermacher/Misal-7B-base-v0.1-GGUF/resolve/main/Misal-7B-base-v0.1.f16.gguf) | f16 | 13.8 | 16 bpw, overkill |
50
 
51
  Here is a handy graph by ikawrakow comparing some lower-quality quant
52
  types (lower is better):