mradermacher commited on
Commit
f7089ac
·
verified ·
1 Parent(s): 6378b10

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +11 -0
README.md CHANGED
@@ -44,8 +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/EHR-R1-8B-i1-GGUF/resolve/main/EHR-R1-8B.imatrix.gguf) | imatrix | 0.1 | imatrix file (for creating your own qwuants) |
 
 
 
47
  | [GGUF](https://huggingface.co/mradermacher/EHR-R1-8B-i1-GGUF/resolve/main/EHR-R1-8B.i1-Q2_K.gguf) | i1-Q2_K | 3.4 | IQ3_XXS probably better |
 
 
48
  | [GGUF](https://huggingface.co/mradermacher/EHR-R1-8B-i1-GGUF/resolve/main/EHR-R1-8B.i1-IQ3_M.gguf) | i1-IQ3_M | 4.0 | |
 
 
 
 
 
 
49
 
50
  Here is a handy graph by ikawrakow comparing some lower-quality quant
51
  types (lower is better):
 
44
  | Link | Type | Size/GB | Notes |
45
  |:-----|:-----|--------:|:------|
46
  | [GGUF](https://huggingface.co/mradermacher/EHR-R1-8B-i1-GGUF/resolve/main/EHR-R1-8B.imatrix.gguf) | imatrix | 0.1 | imatrix file (for creating your own qwuants) |
47
+ | [GGUF](https://huggingface.co/mradermacher/EHR-R1-8B-i1-GGUF/resolve/main/EHR-R1-8B.i1-IQ1_M.gguf) | i1-IQ1_M | 2.4 | mostly desperate |
48
+ | [GGUF](https://huggingface.co/mradermacher/EHR-R1-8B-i1-GGUF/resolve/main/EHR-R1-8B.i1-IQ2_M.gguf) | i1-IQ2_M | 3.2 | |
49
+ | [GGUF](https://huggingface.co/mradermacher/EHR-R1-8B-i1-GGUF/resolve/main/EHR-R1-8B.i1-Q2_K_S.gguf) | i1-Q2_K_S | 3.2 | very low quality |
50
  | [GGUF](https://huggingface.co/mradermacher/EHR-R1-8B-i1-GGUF/resolve/main/EHR-R1-8B.i1-Q2_K.gguf) | i1-Q2_K | 3.4 | IQ3_XXS probably better |
51
+ | [GGUF](https://huggingface.co/mradermacher/EHR-R1-8B-i1-GGUF/resolve/main/EHR-R1-8B.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 3.5 | lower quality |
52
+ | [GGUF](https://huggingface.co/mradermacher/EHR-R1-8B-i1-GGUF/resolve/main/EHR-R1-8B.i1-Q3_K_S.gguf) | i1-Q3_K_S | 3.9 | IQ3_XS probably better |
53
  | [GGUF](https://huggingface.co/mradermacher/EHR-R1-8B-i1-GGUF/resolve/main/EHR-R1-8B.i1-IQ3_M.gguf) | i1-IQ3_M | 4.0 | |
54
+ | [GGUF](https://huggingface.co/mradermacher/EHR-R1-8B-i1-GGUF/resolve/main/EHR-R1-8B.i1-Q3_K_M.gguf) | i1-Q3_K_M | 4.2 | IQ3_S probably better |
55
+ | [GGUF](https://huggingface.co/mradermacher/EHR-R1-8B-i1-GGUF/resolve/main/EHR-R1-8B.i1-IQ4_XS.gguf) | i1-IQ4_XS | 4.7 | |
56
+ | [GGUF](https://huggingface.co/mradermacher/EHR-R1-8B-i1-GGUF/resolve/main/EHR-R1-8B.i1-IQ4_NL.gguf) | i1-IQ4_NL | 4.9 | prefer IQ4_XS |
57
+ | [GGUF](https://huggingface.co/mradermacher/EHR-R1-8B-i1-GGUF/resolve/main/EHR-R1-8B.i1-Q4_K_S.gguf) | i1-Q4_K_S | 4.9 | optimal size/speed/quality |
58
+ | [GGUF](https://huggingface.co/mradermacher/EHR-R1-8B-i1-GGUF/resolve/main/EHR-R1-8B.i1-Q4_K_M.gguf) | i1-Q4_K_M | 5.1 | fast, recommended |
59
+ | [GGUF](https://huggingface.co/mradermacher/EHR-R1-8B-i1-GGUF/resolve/main/EHR-R1-8B.i1-Q6_K.gguf) | i1-Q6_K | 6.8 | practically like static Q6_K |
60
 
61
  Here is a handy graph by ikawrakow comparing some lower-quality quant
62
  types (lower is better):