mradermacher commited on
Commit
cf34d4c
1 Parent(s): 3f402e4

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +65 -0
README.md CHANGED
@@ -1,3 +1,68 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  <!-- ### convert_type: -->
2
  <!-- ### vocab_type: -->
3
  static quants of https://huggingface.co/allknowingroger/TripleMerge-7B-Ties
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model:
3
+ - allknowingroger/MultiverseEx26-7B-slerp
4
+ - allknowingroger/limyClown-7B-slerp
5
+ - allknowingroger/LeeMerge-7B-slerp
6
+ exported_from: allknowingroger/TripleMerge-7B-Ties
7
+ language:
8
+ - en
9
+ library_name: transformers
10
+ quantized_by: mradermacher
11
+ tags:
12
+ - merge
13
+ - mergekit
14
+ - lazymergekit
15
+ - allknowingroger/MultiverseEx26-7B-slerp
16
+ - allknowingroger/limyClown-7B-slerp
17
+ - allknowingroger/LeeMerge-7B-slerp
18
+ ---
19
+ ## About
20
+
21
  <!-- ### convert_type: -->
22
  <!-- ### vocab_type: -->
23
  static quants of https://huggingface.co/allknowingroger/TripleMerge-7B-Ties
24
+
25
+
26
+ <!-- provided-files -->
27
+ 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.
28
+ ## Usage
29
+
30
+ If you are unsure how to use GGUF files, refer to one of [TheBloke's
31
+ READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
32
+ more details, including on how to concatenate multi-part files.
33
+
34
+ ## Provided Quants
35
+
36
+ (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
37
+
38
+ | Link | Type | Size/GB | Notes |
39
+ |:-----|:-----|--------:|:------|
40
+ | [GGUF](https://huggingface.co/mradermacher/TripleMerge-7B-Ties-GGUF/resolve/main/TripleMerge-7B-Ties.Q2_K.gguf) | Q2_K | 3.0 | |
41
+ | [GGUF](https://huggingface.co/mradermacher/TripleMerge-7B-Ties-GGUF/resolve/main/TripleMerge-7B-Ties.IQ3_XS.gguf) | IQ3_XS | 3.3 | |
42
+ | [GGUF](https://huggingface.co/mradermacher/TripleMerge-7B-Ties-GGUF/resolve/main/TripleMerge-7B-Ties.Q3_K_S.gguf) | Q3_K_S | 3.4 | |
43
+ | [GGUF](https://huggingface.co/mradermacher/TripleMerge-7B-Ties-GGUF/resolve/main/TripleMerge-7B-Ties.IQ3_S.gguf) | IQ3_S | 3.4 | beats Q3_K* |
44
+ | [GGUF](https://huggingface.co/mradermacher/TripleMerge-7B-Ties-GGUF/resolve/main/TripleMerge-7B-Ties.Q3_K_M.gguf) | Q3_K_M | 3.8 | lower quality |
45
+ | [GGUF](https://huggingface.co/mradermacher/TripleMerge-7B-Ties-GGUF/resolve/main/TripleMerge-7B-Ties.Q3_K_L.gguf) | Q3_K_L | 4.1 | |
46
+ | [GGUF](https://huggingface.co/mradermacher/TripleMerge-7B-Ties-GGUF/resolve/main/TripleMerge-7B-Ties.Q4_K_S.gguf) | Q4_K_S | 4.4 | fast, recommended |
47
+ | [GGUF](https://huggingface.co/mradermacher/TripleMerge-7B-Ties-GGUF/resolve/main/TripleMerge-7B-Ties.Q4_K_M.gguf) | Q4_K_M | 4.6 | fast, recommended |
48
+ | [GGUF](https://huggingface.co/mradermacher/TripleMerge-7B-Ties-GGUF/resolve/main/TripleMerge-7B-Ties.Q5_K_S.gguf) | Q5_K_S | 5.3 | |
49
+ | [GGUF](https://huggingface.co/mradermacher/TripleMerge-7B-Ties-GGUF/resolve/main/TripleMerge-7B-Ties.Q5_K_M.gguf) | Q5_K_M | 5.4 | |
50
+ | [GGUF](https://huggingface.co/mradermacher/TripleMerge-7B-Ties-GGUF/resolve/main/TripleMerge-7B-Ties.Q6_K.gguf) | Q6_K | 6.2 | very good quality |
51
+ | [GGUF](https://huggingface.co/mradermacher/TripleMerge-7B-Ties-GGUF/resolve/main/TripleMerge-7B-Ties.Q8_0.gguf) | Q8_0 | 7.9 | fast, best quality |
52
+
53
+
54
+ Here is a handy graph by ikawrakow comparing some lower-quality quant
55
+ types (lower is better):
56
+
57
+ ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png)
58
+
59
+ And here are Artefact2's thoughts on the matter:
60
+ https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
61
+
62
+ ## Thanks
63
+
64
+ I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
65
+ me use its servers and providing upgrades to my workstation to enable
66
+ this work in my free time.
67
+
68
+ <!-- end -->