Hugging Face's logo

mradermacher
/
DeepSeek-V3-Base-GGUF

Transformers
English
Inference Endpoints
Model card Files Community
mradermacher commited on
Commit
1305a99
·
verified ·
1 Parent(s): 8ebe935

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +55 -0
README.md CHANGED
@@ -1,6 +1,61 @@
 
 
 
 
 
 
 
 
 
1
  <!-- ### quantize_version: 2 -->
2
  <!-- ### output_tensor_quantised: 1 -->
3
  <!-- ### convert_type: hf -->
4
  <!-- ### vocab_type: -->
5
  <!-- ### tags: -->
6
  static quants of https://huggingface.co/deepseek-ai/DeepSeek-V3-Base
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: deepseek-ai/DeepSeek-V3-Base
3
+ language:
4
+ - en
5
+ library_name: transformers
6
+ quantized_by: mradermacher
7
+ ---
8
+ ## About
9
+
10
  <!-- ### quantize_version: 2 -->
11
  <!-- ### output_tensor_quantised: 1 -->
12
  <!-- ### convert_type: hf -->
13
  <!-- ### vocab_type: -->
14
  <!-- ### tags: -->
15
  static quants of https://huggingface.co/deepseek-ai/DeepSeek-V3-Base
16
+
17
+ <!-- provided-files -->
18
+ weighted/imatrix quants seem not to be available (by me) at this time. If they do not show up a week or so after the static ones, I have probably not planned for them. Feel free to request them by opening a Community Discussion.
19
+ ## Usage
20
+
21
+ If you are unsure how to use GGUF files, refer to one of [TheBloke's
22
+ READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
23
+ more details, including on how to concatenate multi-part files.
24
+
25
+ ## Provided Quants
26
+
27
+ (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
28
+
29
+ | Link | Type | Size/GB | Notes |
30
+ |:-----|:-----|--------:|:------|
31
+ | [P1](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q2_K.gguf.part1of5) [P2](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q2_K.gguf.part2of5) [P3](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q2_K.gguf.part3of5) [P4](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q2_K.gguf.part4of5) [P5](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q2_K.gguf.part5of5) | Q2_K | 244.1 | |
32
+ | [P1](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q3_K_S.gguf.part1of6) [P2](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q3_K_S.gguf.part2of6) [P3](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q3_K_S.gguf.part3of6) [P4](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q3_K_S.gguf.part4of6) [P5](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q3_K_S.gguf.part5of6) [P6](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q3_K_S.gguf.part6of6) | Q3_K_S | 289.2 | |
33
+ | [P1](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q3_K_M.gguf.part1of7) [P2](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q3_K_M.gguf.part2of7) [P3](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q3_K_M.gguf.part3of7) [P4](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q3_K_M.gguf.part4of7) [P5](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q3_K_M.gguf.part5of7) [P6](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q3_K_M.gguf.part6of7) [P7](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q3_K_M.gguf.part7of7) | Q3_K_M | 319.3 | lower quality |
34
+ | [P1](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q3_K_L.gguf.part1of8) [P2](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q3_K_L.gguf.part2of8) [P3](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q3_K_L.gguf.part3of8) [P4](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q3_K_L.gguf.part4of8) [P5](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q3_K_L.gguf.part5of8) [P6](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q3_K_L.gguf.part6of8) [P7](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q3_K_L.gguf.part7of8) [P8](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q3_K_L.gguf.part8of8) | Q3_K_L | 347.5 | |
35
+ | [P1](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q4_K_S.gguf.part1of8) [P2](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q4_K_S.gguf.part2of8) [P3](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q4_K_S.gguf.part3of8) [P4](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q4_K_S.gguf.part4of8) [P5](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q4_K_S.gguf.part5of8) [P6](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q4_K_S.gguf.part6of8) [P7](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q4_K_S.gguf.part7of8) [P8](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q4_K_S.gguf.part8of8) | Q4_K_S | 380.1 | fast, recommended |
36
+ | [P1](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q4_K_M.gguf.part1of9) [P2](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q4_K_M.gguf.part2of9) [P3](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q4_K_M.gguf.part3of9) [P4](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q4_K_M.gguf.part4of9) [P5](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q4_K_M.gguf.part5of9) [P6](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q4_K_M.gguf.part6of9) [P7](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q4_K_M.gguf.part7of9) [P8](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q4_K_M.gguf.part8of9) [P9](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q4_K_M.gguf.part9of9) | Q4_K_M | 404.5 | fast, recommended |
37
+ | [P1](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q5_K_S.gguf.part01of10) [P2](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q5_K_S.gguf.part02of10) [P3](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q5_K_S.gguf.part03of10) [P4](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q5_K_S.gguf.part04of10) [P5](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q5_K_S.gguf.part05of10) [P6](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q5_K_S.gguf.part06of10) [P7](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q5_K_S.gguf.part07of10) [P8](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q5_K_S.gguf.part08of10) [P9](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q5_K_S.gguf.part09of10) [P10](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q5_K_S.gguf.part10of10) | Q5_K_S | 461.9 | |
38
+ | [P1](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q5_K_M.gguf.part01of10) [P2](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q5_K_M.gguf.part02of10) [P3](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q5_K_M.gguf.part03of10) [P4](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q5_K_M.gguf.part04of10) [P5](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q5_K_M.gguf.part05of10) [P6](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q5_K_M.gguf.part06of10) [P7](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q5_K_M.gguf.part07of10) [P8](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q5_K_M.gguf.part08of10) [P9](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q5_K_M.gguf.part09of10) [P10](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q5_K_M.gguf.part10of10) | Q5_K_M | 475.5 | |
39
+ | [P1](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q6_K.gguf.part01of12) [P2](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q6_K.gguf.part02of12) [P3](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q6_K.gguf.part03of12) [P4](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q6_K.gguf.part04of12) [P5](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q6_K.gguf.part05of12) [P6](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q6_K.gguf.part06of12) [P7](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q6_K.gguf.part07of12) [P8](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q6_K.gguf.part08of12) [P9](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q6_K.gguf.part09of12) [P10](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q6_K.gguf.part10of12) [P11](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q6_K.gguf.part11of12) [P12](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q6_K.gguf.part12of12) | Q6_K | 550.9 | very good quality |
40
+ | [P1](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q8_0.gguf.part01of18) [P2](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q8_0.gguf.part02of18) [P3](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q8_0.gguf.part03of18) [P4](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q8_0.gguf.part04of18) [P5](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q8_0.gguf.part05of18) [P6](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q8_0.gguf.part06of18) [P7](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q8_0.gguf.part07of18) [P8](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q8_0.gguf.part08of18) [P9](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q8_0.gguf.part09of18) [P10](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q8_0.gguf.part10of18) [P11](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q8_0.gguf.part11of18) [P12](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q8_0.gguf.part12of18) [P13](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q8_0.gguf.part13of18) [P14](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q8_0.gguf.part14of18) [P15](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q8_0.gguf.part15of18) [P16](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q8_0.gguf.part16of18) [P17](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q8_0.gguf.part17of18) [P18](https://huggingface.co/mradermacher/DeepSeek-V3-Base-GGUF/resolve/main/DeepSeek-V3-Base.Q8_0.gguf.part18of18) | Q8_0 | 713.4 | fast, best quality |
41
+
42
+ Here is a handy graph by ikawrakow comparing some lower-quality quant
43
+ types (lower is better):
44
+
45
+ ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png)
46
+
47
+ And here are Artefact2's thoughts on the matter:
48
+ https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
49
+
50
+ ## FAQ / Model Request
51
+
52
+ See https://huggingface.co/mradermacher/model_requests for some answers to
53
+ questions you might have and/or if you want some other model quantized.
54
+
55
+ ## Thanks
56
+
57
+ I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
58
+ me use its servers and providing upgrades to my workstation to enable
59
+ this work in my free time. Additional thanks to [@nicoboss](https://huggingface.co/nicoboss) for giving me access to his private supercomputer, enabling me to provide many more imatrix quants, at much higher quality, than I would otherwise be able to.
60
+
61
+ <!-- end -->