mradermacher commited on
Commit
27bf39a
·
verified ·
1 Parent(s): e9c5c4b

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +12 -0
README.md CHANGED
@@ -44,7 +44,19 @@ more details, including on how to concatenate multi-part files.
44
  | Link | Type | Size/GB | Notes |
45
  |:-----|:-----|--------:|:------|
46
  | [GGUF](https://huggingface.co/mradermacher/typhoon-ocr-GGUF/resolve/main/typhoon-ocr.mmproj-Q8_0.gguf) | mmproj-Q8_0 | 0.5 | multi-modal supplement |
 
47
  | [GGUF](https://huggingface.co/mradermacher/typhoon-ocr-GGUF/resolve/main/typhoon-ocr.mmproj-f16.gguf) | mmproj-f16 | 0.9 | multi-modal supplement |
 
 
 
 
 
 
 
 
 
 
 
48
 
49
  Here is a handy graph by ikawrakow comparing some lower-quality quant
50
  types (lower is better):
 
44
  | Link | Type | Size/GB | Notes |
45
  |:-----|:-----|--------:|:------|
46
  | [GGUF](https://huggingface.co/mradermacher/typhoon-ocr-GGUF/resolve/main/typhoon-ocr.mmproj-Q8_0.gguf) | mmproj-Q8_0 | 0.5 | multi-modal supplement |
47
+ | [GGUF](https://huggingface.co/mradermacher/typhoon-ocr-GGUF/resolve/main/typhoon-ocr.Q2_K.gguf) | Q2_K | 0.9 | |
48
  | [GGUF](https://huggingface.co/mradermacher/typhoon-ocr-GGUF/resolve/main/typhoon-ocr.mmproj-f16.gguf) | mmproj-f16 | 0.9 | multi-modal supplement |
49
+ | [GGUF](https://huggingface.co/mradermacher/typhoon-ocr-GGUF/resolve/main/typhoon-ocr.Q3_K_S.gguf) | Q3_K_S | 1.0 | |
50
+ | [GGUF](https://huggingface.co/mradermacher/typhoon-ocr-GGUF/resolve/main/typhoon-ocr.Q3_K_M.gguf) | Q3_K_M | 1.0 | lower quality |
51
+ | [GGUF](https://huggingface.co/mradermacher/typhoon-ocr-GGUF/resolve/main/typhoon-ocr.Q3_K_L.gguf) | Q3_K_L | 1.1 | |
52
+ | [GGUF](https://huggingface.co/mradermacher/typhoon-ocr-GGUF/resolve/main/typhoon-ocr.IQ4_XS.gguf) | IQ4_XS | 1.1 | |
53
+ | [GGUF](https://huggingface.co/mradermacher/typhoon-ocr-GGUF/resolve/main/typhoon-ocr.Q4_K_S.gguf) | Q4_K_S | 1.2 | fast, recommended |
54
+ | [GGUF](https://huggingface.co/mradermacher/typhoon-ocr-GGUF/resolve/main/typhoon-ocr.Q4_K_M.gguf) | Q4_K_M | 1.2 | fast, recommended |
55
+ | [GGUF](https://huggingface.co/mradermacher/typhoon-ocr-GGUF/resolve/main/typhoon-ocr.Q5_K_S.gguf) | Q5_K_S | 1.3 | |
56
+ | [GGUF](https://huggingface.co/mradermacher/typhoon-ocr-GGUF/resolve/main/typhoon-ocr.Q5_K_M.gguf) | Q5_K_M | 1.4 | |
57
+ | [GGUF](https://huggingface.co/mradermacher/typhoon-ocr-GGUF/resolve/main/typhoon-ocr.Q6_K.gguf) | Q6_K | 1.5 | very good quality |
58
+ | [GGUF](https://huggingface.co/mradermacher/typhoon-ocr-GGUF/resolve/main/typhoon-ocr.Q8_0.gguf) | Q8_0 | 1.9 | fast, best quality |
59
+ | [GGUF](https://huggingface.co/mradermacher/typhoon-ocr-GGUF/resolve/main/typhoon-ocr.f16.gguf) | f16 | 3.5 | 16 bpw, overkill |
60
 
61
  Here is a handy graph by ikawrakow comparing some lower-quality quant
62
  types (lower is better):