Hugging Face's logo

mradermacher
/
Qwen1.5-72B-Chat-LLaMAfied-GGUF

Transformers
GGUF
English
Chinese
qwen
qwen1.5
qwen2
llama
Inference Endpoints
Model card Files Community
mradermacher commited on
Commit
8988eb3
·
verified ·
1 Parent(s): c738275

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +63 -0
README.md CHANGED
@@ -1 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  static quants of https://huggingface.co/sayhan/Qwen1.5-72B-Chat-LLaMAfied
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - en
4
+ library_name: transformers
5
+ license: other
6
+ license_link: https://github.com/QwenLM/Qwen/blob/main/Tongyi%20Qianwen%20LICENSE%20AGREEMENT
7
+ license_name: qwen
8
+ quantized_by: mradermacher
9
+ tags:
10
+ - qwen
11
+ - qwen1.5
12
+ - qwen2
13
+ - llama
14
+ ---
15
+ ## About
16
+
17
  static quants of https://huggingface.co/sayhan/Qwen1.5-72B-Chat-LLaMAfied
18
+
19
+
20
+ <!-- provided-files -->
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/Qwen1.5-72B-Chat-LLaMAfied-GGUF/resolve/main/Qwen1.5-72B-Chat-LLaMAfied.Q2_K.gguf) | Q2_K | 28.6 | |
34
+ | [GGUF](https://huggingface.co/mradermacher/Qwen1.5-72B-Chat-LLaMAfied-GGUF/resolve/main/Qwen1.5-72B-Chat-LLaMAfied.IQ3_XS.gguf) | IQ3_XS | 31.5 | |
35
+ | [GGUF](https://huggingface.co/mradermacher/Qwen1.5-72B-Chat-LLaMAfied-GGUF/resolve/main/Qwen1.5-72B-Chat-LLaMAfied.IQ3_S.gguf) | IQ3_S | 33.1 | fast, beats Q3_K* |
36
+ | [GGUF](https://huggingface.co/mradermacher/Qwen1.5-72B-Chat-LLaMAfied-GGUF/resolve/main/Qwen1.5-72B-Chat-LLaMAfied.Q3_K_S.gguf) | Q3_K_S | 33.1 | |
37
+ | [GGUF](https://huggingface.co/mradermacher/Qwen1.5-72B-Chat-LLaMAfied-GGUF/resolve/main/Qwen1.5-72B-Chat-LLaMAfied.IQ3_M.gguf) | IQ3_M | 34.8 | |
38
+ | [GGUF](https://huggingface.co/mradermacher/Qwen1.5-72B-Chat-LLaMAfied-GGUF/resolve/main/Qwen1.5-72B-Chat-LLaMAfied.Q3_K_M.gguf) | Q3_K_M | 36.8 | lower quality |
39
+ | [GGUF](https://huggingface.co/mradermacher/Qwen1.5-72B-Chat-LLaMAfied-GGUF/resolve/main/Qwen1.5-72B-Chat-LLaMAfied.Q3_K_L.gguf) | Q3_K_L | 40.1 | |
40
+ | [GGUF](https://huggingface.co/mradermacher/Qwen1.5-72B-Chat-LLaMAfied-GGUF/resolve/main/Qwen1.5-72B-Chat-LLaMAfied.IQ4_XS.gguf) | IQ4_XS | 40.7 | |
41
+ | [GGUF](https://huggingface.co/mradermacher/Qwen1.5-72B-Chat-LLaMAfied-GGUF/resolve/main/Qwen1.5-72B-Chat-LLaMAfied.Q4_K_S.gguf) | Q4_K_S | 42.9 | fast, medium quality |
42
+ | [GGUF](https://huggingface.co/mradermacher/Qwen1.5-72B-Chat-LLaMAfied-GGUF/resolve/main/Qwen1.5-72B-Chat-LLaMAfied.Q4_K_M.gguf) | Q4_K_M | 45.3 | fast, medium quality |
43
+ | [PART 1](https://huggingface.co/mradermacher/Qwen1.5-72B-Chat-LLaMAfied-GGUF/resolve/main/Qwen1.5-72B-Chat-LLaMAfied.Q5_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Qwen1.5-72B-Chat-LLaMAfied-GGUF/resolve/main/Qwen1.5-72B-Chat-LLaMAfied.Q5_K_S.gguf.part2of2) | Q5_K_S | 51.5 | |
44
+ | [PART 1](https://huggingface.co/mradermacher/Qwen1.5-72B-Chat-LLaMAfied-GGUF/resolve/main/Qwen1.5-72B-Chat-LLaMAfied.Q5_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Qwen1.5-72B-Chat-LLaMAfied-GGUF/resolve/main/Qwen1.5-72B-Chat-LLaMAfied.Q5_K_M.gguf.part2of2) | Q5_K_M | 52.9 | |
45
+ | [PART 1](https://huggingface.co/mradermacher/Qwen1.5-72B-Chat-LLaMAfied-GGUF/resolve/main/Qwen1.5-72B-Chat-LLaMAfied.Q6_K.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Qwen1.5-72B-Chat-LLaMAfied-GGUF/resolve/main/Qwen1.5-72B-Chat-LLaMAfied.Q6_K.gguf.part2of2) | Q6_K | 60.9 | very good quality |
46
+ | [PART 1](https://huggingface.co/mradermacher/Qwen1.5-72B-Chat-LLaMAfied-GGUF/resolve/main/Qwen1.5-72B-Chat-LLaMAfied.Q8_0.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Qwen1.5-72B-Chat-LLaMAfied-GGUF/resolve/main/Qwen1.5-72B-Chat-LLaMAfied.Q8_0.gguf.part2of2) | Q8_0 | 78.1 | fast, best quality |
47
+ | [PART 1](https://huggingface.co/mradermacher/Qwen1.5-72B-Chat-LLaMAfied-GGUF/resolve/main/Qwen1.5-72B-Chat-LLaMAfied.SOURCE.gguf.part1of3) [PART 2](https://huggingface.co/mradermacher/Qwen1.5-72B-Chat-LLaMAfied-GGUF/resolve/main/Qwen1.5-72B-Chat-LLaMAfied.SOURCE.gguf.part2of3) [PART 3](https://huggingface.co/mradermacher/Qwen1.5-72B-Chat-LLaMAfied-GGUF/resolve/main/Qwen1.5-72B-Chat-LLaMAfied.SOURCE.gguf.part3of3) | SOURCE | 144.7 | source gguf, only provided when it was hard to come by |
48
+
49
+
50
+ Here is a handy graph by ikawrakow comparing some lower-quality quant
51
+ types (lower is better):
52
+
53
+ ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png)
54
+
55
+ And here are Artefact2's thoughts on the matter:
56
+ https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
57
+
58
+ ## Thanks
59
+
60
+ I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
61
+ me use its servers and providing upgrades to my workstation to enable
62
+ this work in my free time.
63
+
64
+ <!-- end -->