Hugging Face's logo

mradermacher
/
llama-65b-hf-i1-GGUF

Transformers
GGUF
English
Inference Endpoints
Model card Files Community
mradermacher commited on
Commit
ae1a90c
•
1 Parent(s): 8545095

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +16 -1
README.md CHANGED
@@ -14,7 +14,6 @@ quantized_by: mradermacher
14
  <!-- ### vocab_type: -->
15
  weighted/imatrix quants of https://huggingface.co/Enoch/llama-65b-hf
16
 
17
-
18
  <!-- provided-files -->
19
  static quants are available at https://huggingface.co/mradermacher/llama-65b-hf-GGUF
20
  ## Usage
@@ -29,9 +28,25 @@ more details, including on how to concatenate multi-part files.
29
 
30
  | Link | Type | Size/GB | Notes |
31
  |:-----|:-----|--------:|:------|
 
 
 
 
32
  | [GGUF](https://huggingface.co/mradermacher/llama-65b-hf-i1-GGUF/resolve/main/llama-65b-hf.i1-Q2_K.gguf) | i1-Q2_K | 24.2 | IQ3_XXS probably better |
33
  | [GGUF](https://huggingface.co/mradermacher/llama-65b-hf-i1-GGUF/resolve/main/llama-65b-hf.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 24.8 | lower quality |
 
 
 
 
 
 
 
 
34
  | [GGUF](https://huggingface.co/mradermacher/llama-65b-hf-i1-GGUF/resolve/main/llama-65b-hf.i1-Q4_K_S.gguf) | i1-Q4_K_S | 37.2 | optimal size/speed/quality |
 
 
 
 
35
 
36
 
37
  Here is a handy graph by ikawrakow comparing some lower-quality quant
 
14
  <!-- ### vocab_type: -->
15
  weighted/imatrix quants of https://huggingface.co/Enoch/llama-65b-hf
16
 
 
17
  <!-- provided-files -->
18
  static quants are available at https://huggingface.co/mradermacher/llama-65b-hf-GGUF
19
  ## Usage
 
28
 
29
  | Link | Type | Size/GB | Notes |
30
  |:-----|:-----|--------:|:------|
31
+ | [GGUF](https://huggingface.co/mradermacher/llama-65b-hf-i1-GGUF/resolve/main/llama-65b-hf.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 17.6 | |
32
+ | [GGUF](https://huggingface.co/mradermacher/llama-65b-hf-i1-GGUF/resolve/main/llama-65b-hf.i1-IQ2_XS.gguf) | i1-IQ2_XS | 19.4 | |
33
+ | [GGUF](https://huggingface.co/mradermacher/llama-65b-hf-i1-GGUF/resolve/main/llama-65b-hf.i1-IQ2_S.gguf) | i1-IQ2_S | 20.9 | |
34
+ | [GGUF](https://huggingface.co/mradermacher/llama-65b-hf-i1-GGUF/resolve/main/llama-65b-hf.i1-IQ2_M.gguf) | i1-IQ2_M | 22.5 | |
35
  | [GGUF](https://huggingface.co/mradermacher/llama-65b-hf-i1-GGUF/resolve/main/llama-65b-hf.i1-Q2_K.gguf) | i1-Q2_K | 24.2 | IQ3_XXS probably better |
36
  | [GGUF](https://huggingface.co/mradermacher/llama-65b-hf-i1-GGUF/resolve/main/llama-65b-hf.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 24.8 | lower quality |
37
+ | [GGUF](https://huggingface.co/mradermacher/llama-65b-hf-i1-GGUF/resolve/main/llama-65b-hf.i1-IQ3_XS.gguf) | i1-IQ3_XS | 26.7 | |
38
+ | [GGUF](https://huggingface.co/mradermacher/llama-65b-hf-i1-GGUF/resolve/main/llama-65b-hf.i1-IQ3_S.gguf) | i1-IQ3_S | 28.3 | beats Q3_K* |
39
+ | [GGUF](https://huggingface.co/mradermacher/llama-65b-hf-i1-GGUF/resolve/main/llama-65b-hf.i1-Q3_K_S.gguf) | i1-Q3_K_S | 28.3 | IQ3_XS probably better |
40
+ | [GGUF](https://huggingface.co/mradermacher/llama-65b-hf-i1-GGUF/resolve/main/llama-65b-hf.i1-IQ3_M.gguf) | i1-IQ3_M | 29.9 | |
41
+ | [GGUF](https://huggingface.co/mradermacher/llama-65b-hf-i1-GGUF/resolve/main/llama-65b-hf.i1-Q3_K_M.gguf) | i1-Q3_K_M | 31.7 | IQ3_S probably better |
42
+ | [GGUF](https://huggingface.co/mradermacher/llama-65b-hf-i1-GGUF/resolve/main/llama-65b-hf.i1-Q3_K_L.gguf) | i1-Q3_K_L | 34.7 | IQ3_M probably better |
43
+ | [GGUF](https://huggingface.co/mradermacher/llama-65b-hf-i1-GGUF/resolve/main/llama-65b-hf.i1-IQ4_XS.gguf) | i1-IQ4_XS | 34.9 | |
44
+ | [GGUF](https://huggingface.co/mradermacher/llama-65b-hf-i1-GGUF/resolve/main/llama-65b-hf.i1-Q4_0.gguf) | i1-Q4_0 | 37.0 | fast, low quality |
45
  | [GGUF](https://huggingface.co/mradermacher/llama-65b-hf-i1-GGUF/resolve/main/llama-65b-hf.i1-Q4_K_S.gguf) | i1-Q4_K_S | 37.2 | optimal size/speed/quality |
46
+ | [GGUF](https://huggingface.co/mradermacher/llama-65b-hf-i1-GGUF/resolve/main/llama-65b-hf.i1-Q4_K_M.gguf) | i1-Q4_K_M | 39.4 | fast, recommended |
47
+ | [GGUF](https://huggingface.co/mradermacher/llama-65b-hf-i1-GGUF/resolve/main/llama-65b-hf.i1-Q5_K_S.gguf) | i1-Q5_K_S | 45.0 | |
48
+ | [GGUF](https://huggingface.co/mradermacher/llama-65b-hf-i1-GGUF/resolve/main/llama-65b-hf.i1-Q5_K_M.gguf) | i1-Q5_K_M | 46.3 | |
49
+ | [PART 1](https://huggingface.co/mradermacher/llama-65b-hf-i1-GGUF/resolve/main/llama-65b-hf.i1-Q6_K.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/llama-65b-hf-i1-GGUF/resolve/main/llama-65b-hf.i1-Q6_K.gguf.part2of2) | i1-Q6_K | 53.7 | practically like static Q6_K |
50
 
51
 
52
  Here is a handy graph by ikawrakow comparing some lower-quality quant