Hugging Face's logo

mradermacher
/
Qwen2.5-7B-Instruct-Uncensored-i1-GGUF

Transformers
GGUF
Chinese
English
qwen
uncensored
Inference Endpoints
imatrix
conversational
Model card Files Community
mradermacher commited on
Commit
0bbc2ba
•
1 Parent(s): 0fa99d7

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +80 -0
README.md CHANGED
@@ -1,6 +1,86 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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/Orion-zhen/Qwen2.5-7B-Instruct-Uncensored
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: Orion-zhen/Qwen2.5-7B-Instruct-Uncensored
3
+ datasets:
4
+ - NobodyExistsOnTheInternet/ToxicQAFinal
5
+ - anthracite-org/kalo-opus-instruct-22k-no-refusal
6
+ - Orion-zhen/dpo-toxic-zh
7
+ - unalignment/toxic-dpo-v0.2
8
+ - Crystalcareai/Intel-DPO-Pairs-Norefusals
9
+ language:
10
+ - zh
11
+ - en
12
+ library_name: transformers
13
+ license: gpl-3.0
14
+ quantized_by: mradermacher
15
+ tags:
16
+ - qwen
17
+ - uncensored
18
+ ---
19
+ ## About
20
+
21
  <!-- ### quantize_version: 2 -->
22
  <!-- ### output_tensor_quantised: 1 -->
23
  <!-- ### convert_type: hf -->
24
  <!-- ### vocab_type: -->
25
  <!-- ### tags: nicoboss -->
26
  weighted/imatrix quants of https://huggingface.co/Orion-zhen/Qwen2.5-7B-Instruct-Uncensored
27
+
28
+ <!-- provided-files -->
29
+ static quants are available at https://huggingface.co/mradermacher/Qwen2.5-7B-Instruct-Uncensored-GGUF
30
+ ## Usage
31
+
32
+ If you are unsure how to use GGUF files, refer to one of [TheBloke's
33
+ READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
34
+ more details, including on how to concatenate multi-part files.
35
+
36
+ ## Provided Quants
37
+
38
+ (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
39
+
40
+ | Link | Type | Size/GB | Notes |
41
+ |:-----|:-----|--------:|:------|
42
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-7B-Instruct-Uncensored-i1-GGUF/resolve/main/Qwen2.5-7B-Instruct-Uncensored.i1-IQ1_S.gguf) | i1-IQ1_S | 2.0 | for the desperate |
43
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-7B-Instruct-Uncensored-i1-GGUF/resolve/main/Qwen2.5-7B-Instruct-Uncensored.i1-IQ1_M.gguf) | i1-IQ1_M | 2.1 | mostly desperate |
44
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-7B-Instruct-Uncensored-i1-GGUF/resolve/main/Qwen2.5-7B-Instruct-Uncensored.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 2.4 | |
45
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-7B-Instruct-Uncensored-i1-GGUF/resolve/main/Qwen2.5-7B-Instruct-Uncensored.i1-IQ2_XS.gguf) | i1-IQ2_XS | 2.6 | |
46
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-7B-Instruct-Uncensored-i1-GGUF/resolve/main/Qwen2.5-7B-Instruct-Uncensored.i1-IQ2_S.gguf) | i1-IQ2_S | 2.7 | |
47
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-7B-Instruct-Uncensored-i1-GGUF/resolve/main/Qwen2.5-7B-Instruct-Uncensored.i1-IQ2_M.gguf) | i1-IQ2_M | 2.9 | |
48
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-7B-Instruct-Uncensored-i1-GGUF/resolve/main/Qwen2.5-7B-Instruct-Uncensored.i1-Q2_K.gguf) | i1-Q2_K | 3.1 | IQ3_XXS probably better |
49
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-7B-Instruct-Uncensored-i1-GGUF/resolve/main/Qwen2.5-7B-Instruct-Uncensored.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 3.2 | lower quality |
50
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-7B-Instruct-Uncensored-i1-GGUF/resolve/main/Qwen2.5-7B-Instruct-Uncensored.i1-IQ3_XS.gguf) | i1-IQ3_XS | 3.4 | |
51
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-7B-Instruct-Uncensored-i1-GGUF/resolve/main/Qwen2.5-7B-Instruct-Uncensored.i1-Q3_K_S.gguf) | i1-Q3_K_S | 3.6 | IQ3_XS probably better |
52
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-7B-Instruct-Uncensored-i1-GGUF/resolve/main/Qwen2.5-7B-Instruct-Uncensored.i1-IQ3_S.gguf) | i1-IQ3_S | 3.6 | beats Q3_K* |
53
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-7B-Instruct-Uncensored-i1-GGUF/resolve/main/Qwen2.5-7B-Instruct-Uncensored.i1-IQ3_M.gguf) | i1-IQ3_M | 3.7 | |
54
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-7B-Instruct-Uncensored-i1-GGUF/resolve/main/Qwen2.5-7B-Instruct-Uncensored.i1-Q3_K_M.gguf) | i1-Q3_K_M | 3.9 | IQ3_S probably better |
55
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-7B-Instruct-Uncensored-i1-GGUF/resolve/main/Qwen2.5-7B-Instruct-Uncensored.i1-Q3_K_L.gguf) | i1-Q3_K_L | 4.2 | IQ3_M probably better |
56
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-7B-Instruct-Uncensored-i1-GGUF/resolve/main/Qwen2.5-7B-Instruct-Uncensored.i1-IQ4_XS.gguf) | i1-IQ4_XS | 4.3 | |
57
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-7B-Instruct-Uncensored-i1-GGUF/resolve/main/Qwen2.5-7B-Instruct-Uncensored.i1-Q4_0_4_4.gguf) | i1-Q4_0_4_4 | 4.5 | fast on arm, low quality |
58
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-7B-Instruct-Uncensored-i1-GGUF/resolve/main/Qwen2.5-7B-Instruct-Uncensored.i1-Q4_0_4_8.gguf) | i1-Q4_0_4_8 | 4.5 | fast on arm+i8mm, low quality |
59
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-7B-Instruct-Uncensored-i1-GGUF/resolve/main/Qwen2.5-7B-Instruct-Uncensored.i1-Q4_0_8_8.gguf) | i1-Q4_0_8_8 | 4.5 | fast on arm+sve, low quality |
60
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-7B-Instruct-Uncensored-i1-GGUF/resolve/main/Qwen2.5-7B-Instruct-Uncensored.i1-Q4_0.gguf) | i1-Q4_0 | 4.5 | fast, low quality |
61
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-7B-Instruct-Uncensored-i1-GGUF/resolve/main/Qwen2.5-7B-Instruct-Uncensored.i1-Q4_K_S.gguf) | i1-Q4_K_S | 4.6 | optimal size/speed/quality |
62
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-7B-Instruct-Uncensored-i1-GGUF/resolve/main/Qwen2.5-7B-Instruct-Uncensored.i1-Q4_K_M.gguf) | i1-Q4_K_M | 4.8 | fast, recommended |
63
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-7B-Instruct-Uncensored-i1-GGUF/resolve/main/Qwen2.5-7B-Instruct-Uncensored.i1-Q5_K_S.gguf) | i1-Q5_K_S | 5.4 | |
64
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-7B-Instruct-Uncensored-i1-GGUF/resolve/main/Qwen2.5-7B-Instruct-Uncensored.i1-Q5_K_M.gguf) | i1-Q5_K_M | 5.5 | |
65
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-7B-Instruct-Uncensored-i1-GGUF/resolve/main/Qwen2.5-7B-Instruct-Uncensored.i1-Q6_K.gguf) | i1-Q6_K | 6.4 | practically like static Q6_K |
66
+
67
+ Here is a handy graph by ikawrakow comparing some lower-quality quant
68
+ types (lower is better):
69
+
70
+ ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png)
71
+
72
+ And here are Artefact2's thoughts on the matter:
73
+ https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
74
+
75
+ ## FAQ / Model Request
76
+
77
+ See https://huggingface.co/mradermacher/model_requests for some answers to
78
+ questions you might have and/or if you want some other model quantized.
79
+
80
+ ## Thanks
81
+
82
+ I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
83
+ me use its servers and providing upgrades to my workstation to enable
84
+ 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.
85
+
86
+ <!-- end -->