Hugging Face's logo

mradermacher
/
Gemma-3-4b-it-Uncensored-DBL-X-i1-GGUF

Transformers
GGUF
English
Gemma 3
instruct
128k context
Not-For-All-Audiences
all use cases
r rated
x rated
function calling
roleplaying
chat
Uncensored
creative
general usage
problem solving
brainstorming
solve riddles
fiction writing
plot generation
sub-plot generation
story generation
scene continue
storytelling
fiction story
story
writing
fiction
swearing
horror
imatrix
conversational
Model card Files
xet
Community
mradermacher commited on
Commit
027bef7
·
verified ·
1 Parent(s): c2a79fa

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +107 -0
README.md CHANGED
@@ -1,6 +1,113 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  <!-- ### quantize_version: 2 -->
2
  <!-- ### output_tensor_quantised: 1 -->
3
  <!-- ### convert_type: hf -->
4
  <!-- ### vocab_type: -->
5
  <!-- ### tags: nicoboss -->
6
  weighted/imatrix quants of https://huggingface.co/DavidAU/Gemma-3-4b-it-Uncensored-DBL-X
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: DavidAU/Gemma-3-4b-it-Uncensored-DBL-X
3
+ language:
4
+ - en
5
+ library_name: transformers
6
+ license: apache-2.0
7
+ quantized_by: mradermacher
8
+ tags:
9
+ - Gemma 3
10
+ - instruct
11
+ - 128k context
12
+ - not-for-all-audiences
13
+ - all use cases
14
+ - instruct
15
+ - r rated
16
+ - x rated
17
+ - function calling
18
+ - roleplaying
19
+ - chat
20
+ - Uncensored
21
+ - creative
22
+ - 128k context
23
+ - general usage
24
+ - problem solving
25
+ - brainstorming
26
+ - solve riddles
27
+ - general usage
28
+ - problem solving
29
+ - brainstorming
30
+ - solve riddles
31
+ - fiction writing
32
+ - plot generation
33
+ - sub-plot generation
34
+ - fiction writing
35
+ - story generation
36
+ - scene continue
37
+ - storytelling
38
+ - fiction story
39
+ - story
40
+ - writing
41
+ - fiction
42
+ - roleplaying
43
+ - swearing
44
+ - horror
45
+ ---
46
+ ## About
47
+
48
  <!-- ### quantize_version: 2 -->
49
  <!-- ### output_tensor_quantised: 1 -->
50
  <!-- ### convert_type: hf -->
51
  <!-- ### vocab_type: -->
52
  <!-- ### tags: nicoboss -->
53
  weighted/imatrix quants of https://huggingface.co/DavidAU/Gemma-3-4b-it-Uncensored-DBL-X
54
+
55
+ <!-- provided-files -->
56
+ static quants are available at https://huggingface.co/mradermacher/Gemma-3-4b-it-Uncensored-DBL-X-GGUF
57
+ ## Usage
58
+
59
+ If you are unsure how to use GGUF files, refer to one of [TheBloke's
60
+ READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
61
+ more details, including on how to concatenate multi-part files.
62
+
63
+ ## Provided Quants
64
+
65
+ (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
66
+
67
+ | Link | Type | Size/GB | Notes |
68
+ |:-----|:-----|--------:|:------|
69
+ | [GGUF](https://huggingface.co/mradermacher/Gemma-3-4b-it-Uncensored-DBL-X-i1-GGUF/resolve/main/Gemma-3-4b-it-Uncensored-DBL-X.i1-IQ1_S.gguf) | i1-IQ1_S | 1.3 | for the desperate |
70
+ | [GGUF](https://huggingface.co/mradermacher/Gemma-3-4b-it-Uncensored-DBL-X-i1-GGUF/resolve/main/Gemma-3-4b-it-Uncensored-DBL-X.i1-IQ1_M.gguf) | i1-IQ1_M | 1.4 | mostly desperate |
71
+ | [GGUF](https://huggingface.co/mradermacher/Gemma-3-4b-it-Uncensored-DBL-X-i1-GGUF/resolve/main/Gemma-3-4b-it-Uncensored-DBL-X.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 1.5 | |
72
+ | [GGUF](https://huggingface.co/mradermacher/Gemma-3-4b-it-Uncensored-DBL-X-i1-GGUF/resolve/main/Gemma-3-4b-it-Uncensored-DBL-X.i1-IQ2_XS.gguf) | i1-IQ2_XS | 1.6 | |
73
+ | [GGUF](https://huggingface.co/mradermacher/Gemma-3-4b-it-Uncensored-DBL-X-i1-GGUF/resolve/main/Gemma-3-4b-it-Uncensored-DBL-X.i1-IQ2_S.gguf) | i1-IQ2_S | 1.7 | |
74
+ | [GGUF](https://huggingface.co/mradermacher/Gemma-3-4b-it-Uncensored-DBL-X-i1-GGUF/resolve/main/Gemma-3-4b-it-Uncensored-DBL-X.i1-IQ2_M.gguf) | i1-IQ2_M | 1.8 | |
75
+ | [GGUF](https://huggingface.co/mradermacher/Gemma-3-4b-it-Uncensored-DBL-X-i1-GGUF/resolve/main/Gemma-3-4b-it-Uncensored-DBL-X.i1-Q2_K_S.gguf) | i1-Q2_K_S | 1.9 | very low quality |
76
+ | [GGUF](https://huggingface.co/mradermacher/Gemma-3-4b-it-Uncensored-DBL-X-i1-GGUF/resolve/main/Gemma-3-4b-it-Uncensored-DBL-X.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 1.9 | lower quality |
77
+ | [GGUF](https://huggingface.co/mradermacher/Gemma-3-4b-it-Uncensored-DBL-X-i1-GGUF/resolve/main/Gemma-3-4b-it-Uncensored-DBL-X.i1-Q2_K.gguf) | i1-Q2_K | 2.0 | IQ3_XXS probably better |
78
+ | [GGUF](https://huggingface.co/mradermacher/Gemma-3-4b-it-Uncensored-DBL-X-i1-GGUF/resolve/main/Gemma-3-4b-it-Uncensored-DBL-X.i1-IQ3_XS.gguf) | i1-IQ3_XS | 2.1 | |
79
+ | [GGUF](https://huggingface.co/mradermacher/Gemma-3-4b-it-Uncensored-DBL-X-i1-GGUF/resolve/main/Gemma-3-4b-it-Uncensored-DBL-X.i1-IQ3_S.gguf) | i1-IQ3_S | 2.2 | beats Q3_K* |
80
+ | [GGUF](https://huggingface.co/mradermacher/Gemma-3-4b-it-Uncensored-DBL-X-i1-GGUF/resolve/main/Gemma-3-4b-it-Uncensored-DBL-X.i1-Q3_K_S.gguf) | i1-Q3_K_S | 2.2 | IQ3_XS probably better |
81
+ | [GGUF](https://huggingface.co/mradermacher/Gemma-3-4b-it-Uncensored-DBL-X-i1-GGUF/resolve/main/Gemma-3-4b-it-Uncensored-DBL-X.i1-IQ3_M.gguf) | i1-IQ3_M | 2.3 | |
82
+ | [GGUF](https://huggingface.co/mradermacher/Gemma-3-4b-it-Uncensored-DBL-X-i1-GGUF/resolve/main/Gemma-3-4b-it-Uncensored-DBL-X.i1-Q3_K_M.gguf) | i1-Q3_K_M | 2.4 | IQ3_S probably better |
83
+ | [GGUF](https://huggingface.co/mradermacher/Gemma-3-4b-it-Uncensored-DBL-X-i1-GGUF/resolve/main/Gemma-3-4b-it-Uncensored-DBL-X.i1-Q3_K_L.gguf) | i1-Q3_K_L | 2.5 | IQ3_M probably better |
84
+ | [GGUF](https://huggingface.co/mradermacher/Gemma-3-4b-it-Uncensored-DBL-X-i1-GGUF/resolve/main/Gemma-3-4b-it-Uncensored-DBL-X.i1-IQ4_XS.gguf) | i1-IQ4_XS | 2.6 | |
85
+ | [GGUF](https://huggingface.co/mradermacher/Gemma-3-4b-it-Uncensored-DBL-X-i1-GGUF/resolve/main/Gemma-3-4b-it-Uncensored-DBL-X.i1-IQ4_NL.gguf) | i1-IQ4_NL | 2.7 | prefer IQ4_XS |
86
+ | [GGUF](https://huggingface.co/mradermacher/Gemma-3-4b-it-Uncensored-DBL-X-i1-GGUF/resolve/main/Gemma-3-4b-it-Uncensored-DBL-X.i1-Q4_0.gguf) | i1-Q4_0 | 2.7 | fast, low quality |
87
+ | [GGUF](https://huggingface.co/mradermacher/Gemma-3-4b-it-Uncensored-DBL-X-i1-GGUF/resolve/main/Gemma-3-4b-it-Uncensored-DBL-X.i1-Q4_K_S.gguf) | i1-Q4_K_S | 2.7 | optimal size/speed/quality |
88
+ | [GGUF](https://huggingface.co/mradermacher/Gemma-3-4b-it-Uncensored-DBL-X-i1-GGUF/resolve/main/Gemma-3-4b-it-Uncensored-DBL-X.i1-Q4_K_M.gguf) | i1-Q4_K_M | 2.8 | fast, recommended |
89
+ | [GGUF](https://huggingface.co/mradermacher/Gemma-3-4b-it-Uncensored-DBL-X-i1-GGUF/resolve/main/Gemma-3-4b-it-Uncensored-DBL-X.i1-Q4_1.gguf) | i1-Q4_1 | 2.9 | |
90
+ | [GGUF](https://huggingface.co/mradermacher/Gemma-3-4b-it-Uncensored-DBL-X-i1-GGUF/resolve/main/Gemma-3-4b-it-Uncensored-DBL-X.i1-Q5_K_S.gguf) | i1-Q5_K_S | 3.1 | |
91
+ | [GGUF](https://huggingface.co/mradermacher/Gemma-3-4b-it-Uncensored-DBL-X-i1-GGUF/resolve/main/Gemma-3-4b-it-Uncensored-DBL-X.i1-Q5_K_M.gguf) | i1-Q5_K_M | 3.2 | |
92
+ | [GGUF](https://huggingface.co/mradermacher/Gemma-3-4b-it-Uncensored-DBL-X-i1-GGUF/resolve/main/Gemma-3-4b-it-Uncensored-DBL-X.i1-Q6_K.gguf) | i1-Q6_K | 3.6 | practically like static Q6_K |
93
+
94
+ Here is a handy graph by ikawrakow comparing some lower-quality quant
95
+ types (lower is better):
96
+
97
+ ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png)
98
+
99
+ And here are Artefact2's thoughts on the matter:
100
+ https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
101
+
102
+ ## FAQ / Model Request
103
+
104
+ See https://huggingface.co/mradermacher/model_requests for some answers to
105
+ questions you might have and/or if you want some other model quantized.
106
+
107
+ ## Thanks
108
+
109
+ I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
110
+ me use its servers and providing upgrades to my workstation to enable
111
+ 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.
112
+
113
+ <!-- end -->