Hugging Face's logo

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

Translation
Transformers
GGUF
Marathi
marathi
sentiment analysis
reading comprehension
paraphrasing
Inference Endpoints
conversational
Model card Files Community
mradermacher commited on
Commit
6828325
·
verified ·
1 Parent(s): 6e72072

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +6 -0
README.md CHANGED
@@ -44,8 +44,14 @@ more details, including on how to concatenate multi-part files.
44
  | [GGUF](https://huggingface.co/mradermacher/Misal-7B-instruct-v0.1-GGUF/resolve/main/Misal-7B-instruct-v0.1.Q3_K_S.gguf) | Q3_K_S | 3.1 | |
45
  | [GGUF](https://huggingface.co/mradermacher/Misal-7B-instruct-v0.1-GGUF/resolve/main/Misal-7B-instruct-v0.1.Q3_K_M.gguf) | Q3_K_M | 3.5 | lower quality |
46
  | [GGUF](https://huggingface.co/mradermacher/Misal-7B-instruct-v0.1-GGUF/resolve/main/Misal-7B-instruct-v0.1.Q3_K_L.gguf) | Q3_K_L | 3.8 | |
 
47
  | [GGUF](https://huggingface.co/mradermacher/Misal-7B-instruct-v0.1-GGUF/resolve/main/Misal-7B-instruct-v0.1.Q4_K_S.gguf) | Q4_K_S | 4.0 | fast, recommended |
 
 
 
48
  | [GGUF](https://huggingface.co/mradermacher/Misal-7B-instruct-v0.1-GGUF/resolve/main/Misal-7B-instruct-v0.1.Q6_K.gguf) | Q6_K | 5.7 | very good quality |
 
 
49
 
50
  Here is a handy graph by ikawrakow comparing some lower-quality quant
51
  types (lower is better):
 
44
  | [GGUF](https://huggingface.co/mradermacher/Misal-7B-instruct-v0.1-GGUF/resolve/main/Misal-7B-instruct-v0.1.Q3_K_S.gguf) | Q3_K_S | 3.1 | |
45
  | [GGUF](https://huggingface.co/mradermacher/Misal-7B-instruct-v0.1-GGUF/resolve/main/Misal-7B-instruct-v0.1.Q3_K_M.gguf) | Q3_K_M | 3.5 | lower quality |
46
  | [GGUF](https://huggingface.co/mradermacher/Misal-7B-instruct-v0.1-GGUF/resolve/main/Misal-7B-instruct-v0.1.Q3_K_L.gguf) | Q3_K_L | 3.8 | |
47
+ | [GGUF](https://huggingface.co/mradermacher/Misal-7B-instruct-v0.1-GGUF/resolve/main/Misal-7B-instruct-v0.1.IQ4_XS.gguf) | IQ4_XS | 3.8 | |
48
  | [GGUF](https://huggingface.co/mradermacher/Misal-7B-instruct-v0.1-GGUF/resolve/main/Misal-7B-instruct-v0.1.Q4_K_S.gguf) | Q4_K_S | 4.0 | fast, recommended |
49
+ | [GGUF](https://huggingface.co/mradermacher/Misal-7B-instruct-v0.1-GGUF/resolve/main/Misal-7B-instruct-v0.1.Q4_K_M.gguf) | Q4_K_M | 4.3 | fast, recommended |
50
+ | [GGUF](https://huggingface.co/mradermacher/Misal-7B-instruct-v0.1-GGUF/resolve/main/Misal-7B-instruct-v0.1.Q5_K_S.gguf) | Q5_K_S | 4.8 | |
51
+ | [GGUF](https://huggingface.co/mradermacher/Misal-7B-instruct-v0.1-GGUF/resolve/main/Misal-7B-instruct-v0.1.Q5_K_M.gguf) | Q5_K_M | 5.0 | |
52
  | [GGUF](https://huggingface.co/mradermacher/Misal-7B-instruct-v0.1-GGUF/resolve/main/Misal-7B-instruct-v0.1.Q6_K.gguf) | Q6_K | 5.7 | very good quality |
53
+ | [GGUF](https://huggingface.co/mradermacher/Misal-7B-instruct-v0.1-GGUF/resolve/main/Misal-7B-instruct-v0.1.Q8_0.gguf) | Q8_0 | 7.4 | fast, best quality |
54
+ | [GGUF](https://huggingface.co/mradermacher/Misal-7B-instruct-v0.1-GGUF/resolve/main/Misal-7B-instruct-v0.1.f16.gguf) | f16 | 13.8 | 16 bpw, overkill |
55
 
56
  Here is a handy graph by ikawrakow comparing some lower-quality quant
57
  types (lower is better):