Hugging Face's logo

mradermacher
/
Qwen2.5-32B-GGUF

Transformers
GGUF
English
conversational
Model card Files
xet
Community
1

Improve language tag

#1
by lbourdois - opened
base: refs/heads/main
←
from: refs/pr/1
Discussion Files changed
+88
-76
Files changed (1) hide show
  1. README.md +88 -76
README.md CHANGED
@@ -1,76 +1,88 @@
1
- ---
2
- base_model: Qwen/Qwen2.5-32B
3
- language:
4
- - en
5
- library_name: transformers
6
- license: apache-2.0
7
- license_link: https://huggingface.co/Qwen/Qwen2.5-32B/blob/main/LICENSE
8
- quantized_by: mradermacher
9
- ---
10
- ## About
11
-
12
- <!-- ### quantize_version: 2 -->
13
- <!-- ### output_tensor_quantised: 1 -->
14
- <!-- ### convert_type: hf -->
15
- <!-- ### vocab_type: -->
16
- <!-- ### tags: -->
17
- static quants of https://huggingface.co/Qwen/Qwen2.5-32B
18
-
19
- <!-- provided-files -->
20
- weighted/imatrix quants are available at https://huggingface.co/mradermacher/Qwen2.5-32B-i1-GGUF
21
- ## Usage
22
-
23
- If you are unsure how to use GGUF files, refer to one of [TheBloke's
24
- READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
25
- more details, including on how to concatenate multi-part files.
26
-
27
- ## Provided Quants
28
-
29
- (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
30
-
31
- | Link | Type | Size/GB | Notes |
32
- |:-----|:-----|--------:|:------|
33
- | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q2_K.gguf) | Q2_K | 12.4 | |
34
- | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.IQ3_XS.gguf) | IQ3_XS | 13.8 | |
35
- | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q3_K_S.gguf) | Q3_K_S | 14.5 | |
36
- | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.IQ3_S.gguf) | IQ3_S | 14.5 | beats Q3_K* |
37
- | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.IQ3_M.gguf) | IQ3_M | 14.9 | |
38
- | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q3_K_M.gguf) | Q3_K_M | 16.0 | lower quality |
39
- | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q3_K_L.gguf) | Q3_K_L | 17.3 | |
40
- | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.IQ4_XS.gguf) | IQ4_XS | 18.0 | |
41
- | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q4_0.gguf) | Q4_0 | 18.7 | fast, low quality |
42
- | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q4_0_4_4.gguf) | Q4_0_4_4 | 18.7 | fast on arm, low quality |
43
- | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q4_0_4_8.gguf) | Q4_0_4_8 | 18.7 | fast on arm+i8mm, low quality |
44
- | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q4_0_8_8.gguf) | Q4_0_8_8 | 18.7 | fast on arm+sve, low quality |
45
- | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q4_K_S.gguf) | Q4_K_S | 18.9 | fast, recommended |
46
- | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.IQ4_NL.gguf) | IQ4_NL | 18.9 | prefer IQ4_XS |
47
- | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q4_K_M.gguf) | Q4_K_M | 20.0 | fast, recommended |
48
- | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q4_1.gguf) | Q4_1 | 20.7 | |
49
- | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q5_0.gguf) | Q5_0 | 22.7 | |
50
- | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q5_K_S.gguf) | Q5_K_S | 22.7 | |
51
- | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q5_K_M.gguf) | Q5_K_M | 23.4 | |
52
- | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q5_1.gguf) | Q5_1 | 24.7 | |
53
- | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q6_K.gguf) | Q6_K | 27.0 | very good quality |
54
- | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q8_0.gguf) | Q8_0 | 34.9 | fast, best quality |
55
- | [PART 1](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.SOURCE.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.SOURCE.gguf.part2of2) | SOURCE | 65.6 | source gguf, only provided when it was hard to come by |
56
-
57
- Here is a handy graph by ikawrakow comparing some lower-quality quant
58
- types (lower is better):
59
-
60
- ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png)
61
-
62
- And here are Artefact2's thoughts on the matter:
63
- https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
64
-
65
- ## FAQ / Model Request
66
-
67
- See https://huggingface.co/mradermacher/model_requests for some answers to
68
- questions you might have and/or if you want some other model quantized.
69
-
70
- ## Thanks
71
-
72
- I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
73
- me use its servers and providing upgrades to my workstation to enable
74
- 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.
75
-
76
- <!-- end -->
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: Qwen/Qwen2.5-32B
3
+ language:
4
+ - zho
5
+ - eng
6
+ - fra
7
+ - spa
8
+ - por
9
+ - deu
10
+ - ita
11
+ - rus
12
+ - jpn
13
+ - kor
14
+ - vie
15
+ - tha
16
+ - ara
17
+ library_name: transformers
18
+ license: apache-2.0
19
+ license_link: https://huggingface.co/Qwen/Qwen2.5-32B/blob/main/LICENSE
20
+ quantized_by: mradermacher
21
+ ---
22
+ ## About
23
+
24
+ <!-- ### quantize_version: 2 -->
25
+ <!-- ### output_tensor_quantised: 1 -->
26
+ <!-- ### convert_type: hf -->
27
+ <!-- ### vocab_type: -->
28
+ <!-- ### tags: -->
29
+ static quants of https://huggingface.co/Qwen/Qwen2.5-32B
30
+
31
+ <!-- provided-files -->
32
+ weighted/imatrix quants are available at https://huggingface.co/mradermacher/Qwen2.5-32B-i1-GGUF
33
+ ## Usage
34
+
35
+ If you are unsure how to use GGUF files, refer to one of [TheBloke's
36
+ READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
37
+ more details, including on how to concatenate multi-part files.
38
+
39
+ ## Provided Quants
40
+
41
+ (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
42
+
43
+ | Link | Type | Size/GB | Notes |
44
+ |:-----|:-----|--------:|:------|
45
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q2_K.gguf) | Q2_K | 12.4 | |
46
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.IQ3_XS.gguf) | IQ3_XS | 13.8 | |
47
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q3_K_S.gguf) | Q3_K_S | 14.5 | |
48
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.IQ3_S.gguf) | IQ3_S | 14.5 | beats Q3_K* |
49
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.IQ3_M.gguf) | IQ3_M | 14.9 | |
50
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q3_K_M.gguf) | Q3_K_M | 16.0 | lower quality |
51
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q3_K_L.gguf) | Q3_K_L | 17.3 | |
52
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.IQ4_XS.gguf) | IQ4_XS | 18.0 | |
53
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q4_0.gguf) | Q4_0 | 18.7 | fast, low quality |
54
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q4_0_4_4.gguf) | Q4_0_4_4 | 18.7 | fast on arm, low quality |
55
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q4_0_4_8.gguf) | Q4_0_4_8 | 18.7 | fast on arm+i8mm, low quality |
56
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q4_0_8_8.gguf) | Q4_0_8_8 | 18.7 | fast on arm+sve, low quality |
57
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q4_K_S.gguf) | Q4_K_S | 18.9 | fast, recommended |
58
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.IQ4_NL.gguf) | IQ4_NL | 18.9 | prefer IQ4_XS |
59
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q4_K_M.gguf) | Q4_K_M | 20.0 | fast, recommended |
60
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q4_1.gguf) | Q4_1 | 20.7 | |
61
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q5_0.gguf) | Q5_0 | 22.7 | |
62
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q5_K_S.gguf) | Q5_K_S | 22.7 | |
63
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q5_K_M.gguf) | Q5_K_M | 23.4 | |
64
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q5_1.gguf) | Q5_1 | 24.7 | |
65
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q6_K.gguf) | Q6_K | 27.0 | very good quality |
66
+ | [GGUF](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.Q8_0.gguf) | Q8_0 | 34.9 | fast, best quality |
67
+ | [PART 1](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.SOURCE.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Qwen2.5-32B-GGUF/resolve/main/Qwen2.5-32B.SOURCE.gguf.part2of2) | SOURCE | 65.6 | source gguf, only provided when it was hard to come by |
68
+
69
+ Here is a handy graph by ikawrakow comparing some lower-quality quant
70
+ types (lower is better):
71
+
72
+ ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png)
73
+
74
+ And here are Artefact2's thoughts on the matter:
75
+ https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
76
+
77
+ ## FAQ / Model Request
78
+
79
+ See https://huggingface.co/mradermacher/model_requests for some answers to
80
+ questions you might have and/or if you want some other model quantized.
81
+
82
+ ## Thanks
83
+
84
+ I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
85
+ me use its servers and providing upgrades to my workstation to enable
86
+ 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.
87
+
88
+ <!-- end -->