Hugging Face's logo

mradermacher
/
Salamandra-7b-RAG-v2-i1-GGUF

Transformers
GGUF
Inference Endpoints
imatrix
conversational
Model card Files Community
mradermacher commited on
Commit
56a3c2f
•
1 Parent(s): 42e5a26

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +105 -0
README.md CHANGED
@@ -1,6 +1,111 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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/langtech-dev/Salamandra-7b-RAG-v2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: langtech-dev/Salamandra-7b-RAG-v2
3
+ language:
4
+ - bg
5
+ - ca
6
+ - code
7
+ - cs
8
+ - cy
9
+ - da
10
+ - de
11
+ - el
12
+ - en
13
+ - es
14
+ - et
15
+ - eu
16
+ - fi
17
+ - fr
18
+ - ga
19
+ - gl
20
+ - hr
21
+ - hu
22
+ - it
23
+ - lt
24
+ - lv
25
+ - mt
26
+ - nl
27
+ - nn
28
+ - \no
29
+ - oc
30
+ - pl
31
+ - pt
32
+ - ro
33
+ - ru
34
+ - sh
35
+ - sk
36
+ - sl
37
+ - sr
38
+ - sv
39
+ - uk
40
+ library_name: transformers
41
+ license: apache-2.0
42
+ quantized_by: mradermacher
43
+ ---
44
+ ## About
45
+
46
  <!-- ### quantize_version: 2 -->
47
  <!-- ### output_tensor_quantised: 1 -->
48
  <!-- ### convert_type: hf -->
49
  <!-- ### vocab_type: -->
50
  <!-- ### tags: nicoboss -->
51
  weighted/imatrix quants of https://huggingface.co/langtech-dev/Salamandra-7b-RAG-v2
52
+
53
+ <!-- provided-files -->
54
+ static quants are available at https://huggingface.co/mradermacher/Salamandra-7b-RAG-v2-GGUF
55
+ ## Usage
56
+
57
+ If you are unsure how to use GGUF files, refer to one of [TheBloke's
58
+ READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
59
+ more details, including on how to concatenate multi-part files.
60
+
61
+ ## Provided Quants
62
+
63
+ (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
64
+
65
+ | Link | Type | Size/GB | Notes |
66
+ |:-----|:-----|--------:|:------|
67
+ | [GGUF](https://huggingface.co/mradermacher/Salamandra-7b-RAG-v2-i1-GGUF/resolve/main/Salamandra-7b-RAG-v2.i1-IQ1_S.gguf) | i1-IQ1_S | 2.4 | for the desperate |
68
+ | [GGUF](https://huggingface.co/mradermacher/Salamandra-7b-RAG-v2-i1-GGUF/resolve/main/Salamandra-7b-RAG-v2.i1-IQ1_M.gguf) | i1-IQ1_M | 2.5 | mostly desperate |
69
+ | [GGUF](https://huggingface.co/mradermacher/Salamandra-7b-RAG-v2-i1-GGUF/resolve/main/Salamandra-7b-RAG-v2.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 2.7 | |
70
+ | [GGUF](https://huggingface.co/mradermacher/Salamandra-7b-RAG-v2-i1-GGUF/resolve/main/Salamandra-7b-RAG-v2.i1-IQ2_XS.gguf) | i1-IQ2_XS | 2.9 | |
71
+ | [GGUF](https://huggingface.co/mradermacher/Salamandra-7b-RAG-v2-i1-GGUF/resolve/main/Salamandra-7b-RAG-v2.i1-IQ2_S.gguf) | i1-IQ2_S | 3.1 | |
72
+ | [GGUF](https://huggingface.co/mradermacher/Salamandra-7b-RAG-v2-i1-GGUF/resolve/main/Salamandra-7b-RAG-v2.i1-IQ2_M.gguf) | i1-IQ2_M | 3.2 | |
73
+ | [GGUF](https://huggingface.co/mradermacher/Salamandra-7b-RAG-v2-i1-GGUF/resolve/main/Salamandra-7b-RAG-v2.i1-Q2_K.gguf) | i1-Q2_K | 3.4 | IQ3_XXS probably better |
74
+ | [GGUF](https://huggingface.co/mradermacher/Salamandra-7b-RAG-v2-i1-GGUF/resolve/main/Salamandra-7b-RAG-v2.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 3.5 | lower quality |
75
+ | [GGUF](https://huggingface.co/mradermacher/Salamandra-7b-RAG-v2-i1-GGUF/resolve/main/Salamandra-7b-RAG-v2.i1-IQ3_XS.gguf) | i1-IQ3_XS | 3.7 | |
76
+ | [GGUF](https://huggingface.co/mradermacher/Salamandra-7b-RAG-v2-i1-GGUF/resolve/main/Salamandra-7b-RAG-v2.i1-Q3_K_S.gguf) | i1-Q3_K_S | 3.9 | IQ3_XS probably better |
77
+ | [GGUF](https://huggingface.co/mradermacher/Salamandra-7b-RAG-v2-i1-GGUF/resolve/main/Salamandra-7b-RAG-v2.i1-IQ3_S.gguf) | i1-IQ3_S | 3.9 | beats Q3_K* |
78
+ | [GGUF](https://huggingface.co/mradermacher/Salamandra-7b-RAG-v2-i1-GGUF/resolve/main/Salamandra-7b-RAG-v2.i1-IQ3_M.gguf) | i1-IQ3_M | 4.0 | |
79
+ | [GGUF](https://huggingface.co/mradermacher/Salamandra-7b-RAG-v2-i1-GGUF/resolve/main/Salamandra-7b-RAG-v2.i1-Q3_K_M.gguf) | i1-Q3_K_M | 4.1 | IQ3_S probably better |
80
+ | [GGUF](https://huggingface.co/mradermacher/Salamandra-7b-RAG-v2-i1-GGUF/resolve/main/Salamandra-7b-RAG-v2.i1-Q3_K_L.gguf) | i1-Q3_K_L | 4.4 | IQ3_M probably better |
81
+ | [GGUF](https://huggingface.co/mradermacher/Salamandra-7b-RAG-v2-i1-GGUF/resolve/main/Salamandra-7b-RAG-v2.i1-IQ4_XS.gguf) | i1-IQ4_XS | 4.6 | |
82
+ | [GGUF](https://huggingface.co/mradermacher/Salamandra-7b-RAG-v2-i1-GGUF/resolve/main/Salamandra-7b-RAG-v2.i1-Q4_0_4_4.gguf) | i1-Q4_0_4_4 | 4.7 | fast on arm, low quality |
83
+ | [GGUF](https://huggingface.co/mradermacher/Salamandra-7b-RAG-v2-i1-GGUF/resolve/main/Salamandra-7b-RAG-v2.i1-Q4_0_4_8.gguf) | i1-Q4_0_4_8 | 4.7 | fast on arm+i8mm, low quality |
84
+ | [GGUF](https://huggingface.co/mradermacher/Salamandra-7b-RAG-v2-i1-GGUF/resolve/main/Salamandra-7b-RAG-v2.i1-Q4_0_8_8.gguf) | i1-Q4_0_8_8 | 4.7 | fast on arm+sve, low quality |
85
+ | [GGUF](https://huggingface.co/mradermacher/Salamandra-7b-RAG-v2-i1-GGUF/resolve/main/Salamandra-7b-RAG-v2.i1-Q4_0.gguf) | i1-Q4_0 | 4.8 | fast, low quality |
86
+ | [GGUF](https://huggingface.co/mradermacher/Salamandra-7b-RAG-v2-i1-GGUF/resolve/main/Salamandra-7b-RAG-v2.i1-Q4_K_S.gguf) | i1-Q4_K_S | 4.8 | optimal size/speed/quality |
87
+ | [GGUF](https://huggingface.co/mradermacher/Salamandra-7b-RAG-v2-i1-GGUF/resolve/main/Salamandra-7b-RAG-v2.i1-Q4_K_M.gguf) | i1-Q4_K_M | 5.0 | fast, recommended |
88
+ | [GGUF](https://huggingface.co/mradermacher/Salamandra-7b-RAG-v2-i1-GGUF/resolve/main/Salamandra-7b-RAG-v2.i1-Q5_K_S.gguf) | i1-Q5_K_S | 5.6 | |
89
+ | [GGUF](https://huggingface.co/mradermacher/Salamandra-7b-RAG-v2-i1-GGUF/resolve/main/Salamandra-7b-RAG-v2.i1-Q5_K_M.gguf) | i1-Q5_K_M | 5.7 | |
90
+ | [GGUF](https://huggingface.co/mradermacher/Salamandra-7b-RAG-v2-i1-GGUF/resolve/main/Salamandra-7b-RAG-v2.i1-Q6_K.gguf) | i1-Q6_K | 6.5 | practically like static Q6_K |
91
+
92
+ Here is a handy graph by ikawrakow comparing some lower-quality quant
93
+ types (lower is better):
94
+
95
+ ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png)
96
+
97
+ And here are Artefact2's thoughts on the matter:
98
+ https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
99
+
100
+ ## FAQ / Model Request
101
+
102
+ See https://huggingface.co/mradermacher/model_requests for some answers to
103
+ questions you might have and/or if you want some other model quantized.
104
+
105
+ ## Thanks
106
+
107
+ I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
108
+ me use its servers and providing upgrades to my workstation to enable
109
+ 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.
110
+
111
+ <!-- end -->