Hugging Face's logo

mradermacher
/
deepseek-llm-67b-base-i1-GGUF

Transformers
GGUF
English
Inference Endpoints
Model card Files Community
mradermacher commited on
Commit
3e96a16
•
1 Parent(s): dd0aa1d

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +6 -1
README.md CHANGED
@@ -16,7 +16,6 @@ quantized_by: mradermacher
16
  <!-- ### vocab_type: -->
17
  weighted/imatrix quants of https://huggingface.co/deepseek-ai/deepseek-llm-67b-base
18
 
19
-
20
  <!-- provided-files -->
21
  static quants are available at https://huggingface.co/mradermacher/deepseek-llm-67b-base-GGUF
22
  ## Usage
@@ -34,12 +33,18 @@ more details, including on how to concatenate multi-part files.
34
  | [GGUF](https://huggingface.co/mradermacher/deepseek-llm-67b-base-i1-GGUF/resolve/main/deepseek-llm-67b-base.i1-IQ2_M.gguf) | i1-IQ2_M | 23.2 | |
35
  | [GGUF](https://huggingface.co/mradermacher/deepseek-llm-67b-base-i1-GGUF/resolve/main/deepseek-llm-67b-base.i1-Q2_K.gguf) | i1-Q2_K | 25.2 | IQ3_XXS probably better |
36
  | [GGUF](https://huggingface.co/mradermacher/deepseek-llm-67b-base-i1-GGUF/resolve/main/deepseek-llm-67b-base.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 26.2 | lower quality |
 
37
  | [GGUF](https://huggingface.co/mradermacher/deepseek-llm-67b-base-i1-GGUF/resolve/main/deepseek-llm-67b-base.i1-Q3_K_S.gguf) | i1-Q3_K_S | 29.4 | IQ3_XS probably better |
 
 
38
  | [GGUF](https://huggingface.co/mradermacher/deepseek-llm-67b-base-i1-GGUF/resolve/main/deepseek-llm-67b-base.i1-Q3_K_M.gguf) | i1-Q3_K_M | 32.8 | IQ3_S probably better |
39
  | [GGUF](https://huggingface.co/mradermacher/deepseek-llm-67b-base-i1-GGUF/resolve/main/deepseek-llm-67b-base.i1-Q3_K_L.gguf) | i1-Q3_K_L | 35.7 | IQ3_M probably better |
40
  | [GGUF](https://huggingface.co/mradermacher/deepseek-llm-67b-base-i1-GGUF/resolve/main/deepseek-llm-67b-base.i1-IQ4_XS.gguf) | i1-IQ4_XS | 36.3 | |
 
41
  | [GGUF](https://huggingface.co/mradermacher/deepseek-llm-67b-base-i1-GGUF/resolve/main/deepseek-llm-67b-base.i1-Q4_K_S.gguf) | i1-Q4_K_S | 38.5 | optimal size/speed/quality |
42
  | [GGUF](https://huggingface.co/mradermacher/deepseek-llm-67b-base-i1-GGUF/resolve/main/deepseek-llm-67b-base.i1-Q4_K_M.gguf) | i1-Q4_K_M | 40.5 | fast, recommended |
 
 
43
  | [PART 1](https://huggingface.co/mradermacher/deepseek-llm-67b-base-i1-GGUF/resolve/main/deepseek-llm-67b-base.i1-Q6_K.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/deepseek-llm-67b-base-i1-GGUF/resolve/main/deepseek-llm-67b-base.i1-Q6_K.gguf.part2of2) | i1-Q6_K | 55.4 | practically like static Q6_K |
44
 
45
 
 
16
  <!-- ### vocab_type: -->
17
  weighted/imatrix quants of https://huggingface.co/deepseek-ai/deepseek-llm-67b-base
18
 
 
19
  <!-- provided-files -->
20
  static quants are available at https://huggingface.co/mradermacher/deepseek-llm-67b-base-GGUF
21
  ## Usage
 
33
  | [GGUF](https://huggingface.co/mradermacher/deepseek-llm-67b-base-i1-GGUF/resolve/main/deepseek-llm-67b-base.i1-IQ2_M.gguf) | i1-IQ2_M | 23.2 | |
34
  | [GGUF](https://huggingface.co/mradermacher/deepseek-llm-67b-base-i1-GGUF/resolve/main/deepseek-llm-67b-base.i1-Q2_K.gguf) | i1-Q2_K | 25.2 | IQ3_XXS probably better |
35
  | [GGUF](https://huggingface.co/mradermacher/deepseek-llm-67b-base-i1-GGUF/resolve/main/deepseek-llm-67b-base.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 26.2 | lower quality |
36
+ | [GGUF](https://huggingface.co/mradermacher/deepseek-llm-67b-base-i1-GGUF/resolve/main/deepseek-llm-67b-base.i1-IQ3_XS.gguf) | i1-IQ3_XS | 28.0 | |
37
  | [GGUF](https://huggingface.co/mradermacher/deepseek-llm-67b-base-i1-GGUF/resolve/main/deepseek-llm-67b-base.i1-Q3_K_S.gguf) | i1-Q3_K_S | 29.4 | IQ3_XS probably better |
38
+ | [GGUF](https://huggingface.co/mradermacher/deepseek-llm-67b-base-i1-GGUF/resolve/main/deepseek-llm-67b-base.i1-IQ3_S.gguf) | i1-IQ3_S | 29.5 | beats Q3_K* |
39
+ | [GGUF](https://huggingface.co/mradermacher/deepseek-llm-67b-base-i1-GGUF/resolve/main/deepseek-llm-67b-base.i1-IQ3_M.gguf) | i1-IQ3_M | 30.6 | |
40
  | [GGUF](https://huggingface.co/mradermacher/deepseek-llm-67b-base-i1-GGUF/resolve/main/deepseek-llm-67b-base.i1-Q3_K_M.gguf) | i1-Q3_K_M | 32.8 | IQ3_S probably better |
41
  | [GGUF](https://huggingface.co/mradermacher/deepseek-llm-67b-base-i1-GGUF/resolve/main/deepseek-llm-67b-base.i1-Q3_K_L.gguf) | i1-Q3_K_L | 35.7 | IQ3_M probably better |
42
  | [GGUF](https://huggingface.co/mradermacher/deepseek-llm-67b-base-i1-GGUF/resolve/main/deepseek-llm-67b-base.i1-IQ4_XS.gguf) | i1-IQ4_XS | 36.3 | |
43
+ | [GGUF](https://huggingface.co/mradermacher/deepseek-llm-67b-base-i1-GGUF/resolve/main/deepseek-llm-67b-base.i1-Q4_0.gguf) | i1-Q4_0 | 38.4 | fast, low quality |
44
  | [GGUF](https://huggingface.co/mradermacher/deepseek-llm-67b-base-i1-GGUF/resolve/main/deepseek-llm-67b-base.i1-Q4_K_S.gguf) | i1-Q4_K_S | 38.5 | optimal size/speed/quality |
45
  | [GGUF](https://huggingface.co/mradermacher/deepseek-llm-67b-base-i1-GGUF/resolve/main/deepseek-llm-67b-base.i1-Q4_K_M.gguf) | i1-Q4_K_M | 40.5 | fast, recommended |
46
+ | [GGUF](https://huggingface.co/mradermacher/deepseek-llm-67b-base-i1-GGUF/resolve/main/deepseek-llm-67b-base.i1-Q5_K_S.gguf) | i1-Q5_K_S | 46.6 | |
47
+ | [GGUF](https://huggingface.co/mradermacher/deepseek-llm-67b-base-i1-GGUF/resolve/main/deepseek-llm-67b-base.i1-Q5_K_M.gguf) | i1-Q5_K_M | 47.8 | |
48
  | [PART 1](https://huggingface.co/mradermacher/deepseek-llm-67b-base-i1-GGUF/resolve/main/deepseek-llm-67b-base.i1-Q6_K.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/deepseek-llm-67b-base-i1-GGUF/resolve/main/deepseek-llm-67b-base.i1-Q6_K.gguf.part2of2) | i1-Q6_K | 55.4 | practically like static Q6_K |
49
 
50