Hugging Face's logo

mradermacher
/
Teuken-7B-instruct-v0.6-i1-GGUF

Transformers
GGUF
imatrix
Model card Files
xet
Community
1
mradermacher commited on
Commit
3ca71d3
·
verified ·
1 Parent(s): 660518e

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +99 -0
README.md CHANGED
@@ -1,3 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  <!-- ### quantize_version: 2 -->
2
  <!-- ### output_tensor_quantised: 1 -->
3
  <!-- ### convert_type: hf -->
@@ -7,3 +42,67 @@
7
  <!-- ### quants_skip: -->
8
  <!-- ### skip_mmproj: -->
9
  weighted/imatrix quants of https://huggingface.co/openGPT-X/Teuken-7B-instruct-v0.6
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: openGPT-X/Teuken-7B-instruct-v0.6
3
+ language:
4
+ - de
5
+ - bg
6
+ - cs
7
+ - da
8
+ - el
9
+ - en
10
+ - es
11
+ - et
12
+ - fi
13
+ - fr
14
+ - ga
15
+ - hr
16
+ - hu
17
+ - it
18
+ - lt
19
+ - lv
20
+ - mt
21
+ - nl
22
+ - pl
23
+ - pt
24
+ - ro
25
+ - sl
26
+ - sv
27
+ - sk
28
+ library_name: transformers
29
+ license: cc-by-nc-4.0
30
+ mradermacher:
31
+ readme_rev: 1
32
+ quantized_by: mradermacher
33
+ ---
34
+ ## About
35
+
36
  <!-- ### quantize_version: 2 -->
37
  <!-- ### output_tensor_quantised: 1 -->
38
  <!-- ### convert_type: hf -->
 
42
  <!-- ### quants_skip: -->
43
  <!-- ### skip_mmproj: -->
44
  weighted/imatrix quants of https://huggingface.co/openGPT-X/Teuken-7B-instruct-v0.6
45
+
46
+ <!-- provided-files -->
47
+
48
+ ***For a convenient overview and download list, visit our [model page for this model](https://hf.tst.eu/model#Teuken-7B-instruct-v0.6-i1-GGUF).***
49
+
50
+ static quants are available at https://huggingface.co/mradermacher/Teuken-7B-instruct-v0.6-GGUF
51
+ ## Usage
52
+
53
+ If you are unsure how to use GGUF files, refer to one of [TheBloke's
54
+ READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
55
+ more details, including on how to concatenate multi-part files.
56
+
57
+ ## Provided Quants
58
+
59
+ (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
60
+
61
+ | Link | Type | Size/GB | Notes |
62
+ |:-----|:-----|--------:|:------|
63
+ | [GGUF](https://huggingface.co/mradermacher/Teuken-7B-instruct-v0.6-i1-GGUF/resolve/main/Teuken-7B-instruct-v0.6.imatrix.gguf) | imatrix | 0.1 | imatrix file (for creating your own qwuants) |
64
+ | [GGUF](https://huggingface.co/mradermacher/Teuken-7B-instruct-v0.6-i1-GGUF/resolve/main/Teuken-7B-instruct-v0.6.i1-IQ1_S.gguf) | i1-IQ1_S | 2.8 | for the desperate |
65
+ | [GGUF](https://huggingface.co/mradermacher/Teuken-7B-instruct-v0.6-i1-GGUF/resolve/main/Teuken-7B-instruct-v0.6.i1-IQ1_M.gguf) | i1-IQ1_M | 2.9 | mostly desperate |
66
+ | [GGUF](https://huggingface.co/mradermacher/Teuken-7B-instruct-v0.6-i1-GGUF/resolve/main/Teuken-7B-instruct-v0.6.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 3.0 | |
67
+ | [GGUF](https://huggingface.co/mradermacher/Teuken-7B-instruct-v0.6-i1-GGUF/resolve/main/Teuken-7B-instruct-v0.6.i1-IQ2_XS.gguf) | i1-IQ2_XS | 3.2 | |
68
+ | [GGUF](https://huggingface.co/mradermacher/Teuken-7B-instruct-v0.6-i1-GGUF/resolve/main/Teuken-7B-instruct-v0.6.i1-IQ2_S.gguf) | i1-IQ2_S | 3.2 | |
69
+ | [GGUF](https://huggingface.co/mradermacher/Teuken-7B-instruct-v0.6-i1-GGUF/resolve/main/Teuken-7B-instruct-v0.6.i1-IQ2_M.gguf) | i1-IQ2_M | 3.4 | |
70
+ | [GGUF](https://huggingface.co/mradermacher/Teuken-7B-instruct-v0.6-i1-GGUF/resolve/main/Teuken-7B-instruct-v0.6.i1-Q2_K_S.gguf) | i1-Q2_K_S | 3.5 | very low quality |
71
+ | [GGUF](https://huggingface.co/mradermacher/Teuken-7B-instruct-v0.6-i1-GGUF/resolve/main/Teuken-7B-instruct-v0.6.i1-Q2_K.gguf) | i1-Q2_K | 3.5 | IQ3_XXS probably better |
72
+ | [GGUF](https://huggingface.co/mradermacher/Teuken-7B-instruct-v0.6-i1-GGUF/resolve/main/Teuken-7B-instruct-v0.6.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 3.6 | lower quality |
73
+ | [GGUF](https://huggingface.co/mradermacher/Teuken-7B-instruct-v0.6-i1-GGUF/resolve/main/Teuken-7B-instruct-v0.6.i1-IQ3_XS.gguf) | i1-IQ3_XS | 3.8 | |
74
+ | [GGUF](https://huggingface.co/mradermacher/Teuken-7B-instruct-v0.6-i1-GGUF/resolve/main/Teuken-7B-instruct-v0.6.i1-Q3_K_S.gguf) | i1-Q3_K_S | 3.9 | IQ3_XS probably better |
75
+ | [GGUF](https://huggingface.co/mradermacher/Teuken-7B-instruct-v0.6-i1-GGUF/resolve/main/Teuken-7B-instruct-v0.6.i1-IQ3_S.gguf) | i1-IQ3_S | 3.9 | beats Q3_K* |
76
+ | [GGUF](https://huggingface.co/mradermacher/Teuken-7B-instruct-v0.6-i1-GGUF/resolve/main/Teuken-7B-instruct-v0.6.i1-IQ3_M.gguf) | i1-IQ3_M | 4.0 | |
77
+ | [GGUF](https://huggingface.co/mradermacher/Teuken-7B-instruct-v0.6-i1-GGUF/resolve/main/Teuken-7B-instruct-v0.6.i1-Q3_K_M.gguf) | i1-Q3_K_M | 4.2 | IQ3_S probably better |
78
+ | [GGUF](https://huggingface.co/mradermacher/Teuken-7B-instruct-v0.6-i1-GGUF/resolve/main/Teuken-7B-instruct-v0.6.i1-Q3_K_L.gguf) | i1-Q3_K_L | 4.4 | IQ3_M probably better |
79
+ | [GGUF](https://huggingface.co/mradermacher/Teuken-7B-instruct-v0.6-i1-GGUF/resolve/main/Teuken-7B-instruct-v0.6.i1-IQ4_XS.gguf) | i1-IQ4_XS | 4.4 | |
80
+ | [GGUF](https://huggingface.co/mradermacher/Teuken-7B-instruct-v0.6-i1-GGUF/resolve/main/Teuken-7B-instruct-v0.6.i1-IQ4_NL.gguf) | i1-IQ4_NL | 4.6 | prefer IQ4_XS |
81
+ | [GGUF](https://huggingface.co/mradermacher/Teuken-7B-instruct-v0.6-i1-GGUF/resolve/main/Teuken-7B-instruct-v0.6.i1-Q4_0.gguf) | i1-Q4_0 | 4.6 | fast, low quality |
82
+ | [GGUF](https://huggingface.co/mradermacher/Teuken-7B-instruct-v0.6-i1-GGUF/resolve/main/Teuken-7B-instruct-v0.6.i1-Q4_K_S.gguf) | i1-Q4_K_S | 4.8 | optimal size/speed/quality |
83
+ | [GGUF](https://huggingface.co/mradermacher/Teuken-7B-instruct-v0.6-i1-GGUF/resolve/main/Teuken-7B-instruct-v0.6.i1-Q4_1.gguf) | i1-Q4_1 | 5.0 | |
84
+ | [GGUF](https://huggingface.co/mradermacher/Teuken-7B-instruct-v0.6-i1-GGUF/resolve/main/Teuken-7B-instruct-v0.6.i1-Q4_K_M.gguf) | i1-Q4_K_M | 5.1 | fast, recommended |
85
+ | [GGUF](https://huggingface.co/mradermacher/Teuken-7B-instruct-v0.6-i1-GGUF/resolve/main/Teuken-7B-instruct-v0.6.i1-Q5_K_S.gguf) | i1-Q5_K_S | 5.5 | |
86
+ | [GGUF](https://huggingface.co/mradermacher/Teuken-7B-instruct-v0.6-i1-GGUF/resolve/main/Teuken-7B-instruct-v0.6.i1-Q5_K_M.gguf) | i1-Q5_K_M | 5.8 | |
87
+ | [GGUF](https://huggingface.co/mradermacher/Teuken-7B-instruct-v0.6-i1-GGUF/resolve/main/Teuken-7B-instruct-v0.6.i1-Q6_K.gguf) | i1-Q6_K | 6.6 | practically like static Q6_K |
88
+
89
+ Here is a handy graph by ikawrakow comparing some lower-quality quant
90
+ types (lower is better):
91
+
92
+ ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png)
93
+
94
+ And here are Artefact2's thoughts on the matter:
95
+ https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
96
+
97
+ ## FAQ / Model Request
98
+
99
+ See https://huggingface.co/mradermacher/model_requests for some answers to
100
+ questions you might have and/or if you want some other model quantized.
101
+
102
+ ## Thanks
103
+
104
+ I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
105
+ me use its servers and providing upgrades to my workstation to enable
106
+ 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.
107
+
108
+ <!-- end -->