mradermacher commited on
Commit
da54f96
1 Parent(s): 594180d

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +73 -0
README.md CHANGED
@@ -1,6 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
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/grimjim/Magot-v1-Gemma2-8k-9B
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: grimjim/Magot-v1-Gemma2-8k-9B
3
+ language:
4
+ - en
5
+ library_name: transformers
6
+ license: gemma
7
+ quantized_by: mradermacher
8
+ tags:
9
+ - mergekit
10
+ - merge
11
+ ---
12
+ ## About
13
+
14
  <!-- ### quantize_version: 2 -->
15
  <!-- ### output_tensor_quantised: 1 -->
16
  <!-- ### convert_type: hf -->
17
  <!-- ### vocab_type: -->
18
  <!-- ### tags: nicoboss -->
19
  weighted/imatrix quants of https://huggingface.co/grimjim/Magot-v1-Gemma2-8k-9B
20
+
21
+ <!-- provided-files -->
22
+ static quants are available at https://huggingface.co/mradermacher/Magot-v1-Gemma2-8k-9B-GGUF
23
+ ## Usage
24
+
25
+ If you are unsure how to use GGUF files, refer to one of [TheBloke's
26
+ READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
27
+ more details, including on how to concatenate multi-part files.
28
+
29
+ ## Provided Quants
30
+
31
+ (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
32
+
33
+ | Link | Type | Size/GB | Notes |
34
+ |:-----|:-----|--------:|:------|
35
+ | [GGUF](https://huggingface.co/mradermacher/Magot-v1-Gemma2-8k-9B-i1-GGUF/resolve/main/Magot-v1-Gemma2-8k-9B.i1-IQ1_S.gguf) | i1-IQ1_S | 2.5 | for the desperate |
36
+ | [GGUF](https://huggingface.co/mradermacher/Magot-v1-Gemma2-8k-9B-i1-GGUF/resolve/main/Magot-v1-Gemma2-8k-9B.i1-IQ1_M.gguf) | i1-IQ1_M | 2.6 | mostly desperate |
37
+ | [GGUF](https://huggingface.co/mradermacher/Magot-v1-Gemma2-8k-9B-i1-GGUF/resolve/main/Magot-v1-Gemma2-8k-9B.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 2.9 | |
38
+ | [GGUF](https://huggingface.co/mradermacher/Magot-v1-Gemma2-8k-9B-i1-GGUF/resolve/main/Magot-v1-Gemma2-8k-9B.i1-IQ2_XS.gguf) | i1-IQ2_XS | 3.2 | |
39
+ | [GGUF](https://huggingface.co/mradermacher/Magot-v1-Gemma2-8k-9B-i1-GGUF/resolve/main/Magot-v1-Gemma2-8k-9B.i1-IQ2_S.gguf) | i1-IQ2_S | 3.3 | |
40
+ | [GGUF](https://huggingface.co/mradermacher/Magot-v1-Gemma2-8k-9B-i1-GGUF/resolve/main/Magot-v1-Gemma2-8k-9B.i1-IQ2_M.gguf) | i1-IQ2_M | 3.5 | |
41
+ | [GGUF](https://huggingface.co/mradermacher/Magot-v1-Gemma2-8k-9B-i1-GGUF/resolve/main/Magot-v1-Gemma2-8k-9B.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 3.9 | lower quality |
42
+ | [GGUF](https://huggingface.co/mradermacher/Magot-v1-Gemma2-8k-9B-i1-GGUF/resolve/main/Magot-v1-Gemma2-8k-9B.i1-Q2_K.gguf) | i1-Q2_K | 3.9 | IQ3_XXS probably better |
43
+ | [GGUF](https://huggingface.co/mradermacher/Magot-v1-Gemma2-8k-9B-i1-GGUF/resolve/main/Magot-v1-Gemma2-8k-9B.i1-IQ3_XS.gguf) | i1-IQ3_XS | 4.2 | |
44
+ | [GGUF](https://huggingface.co/mradermacher/Magot-v1-Gemma2-8k-9B-i1-GGUF/resolve/main/Magot-v1-Gemma2-8k-9B.i1-IQ3_S.gguf) | i1-IQ3_S | 4.4 | beats Q3_K* |
45
+ | [GGUF](https://huggingface.co/mradermacher/Magot-v1-Gemma2-8k-9B-i1-GGUF/resolve/main/Magot-v1-Gemma2-8k-9B.i1-Q3_K_S.gguf) | i1-Q3_K_S | 4.4 | IQ3_XS probably better |
46
+ | [GGUF](https://huggingface.co/mradermacher/Magot-v1-Gemma2-8k-9B-i1-GGUF/resolve/main/Magot-v1-Gemma2-8k-9B.i1-IQ3_M.gguf) | i1-IQ3_M | 4.6 | |
47
+ | [GGUF](https://huggingface.co/mradermacher/Magot-v1-Gemma2-8k-9B-i1-GGUF/resolve/main/Magot-v1-Gemma2-8k-9B.i1-Q3_K_M.gguf) | i1-Q3_K_M | 4.9 | IQ3_S probably better |
48
+ | [GGUF](https://huggingface.co/mradermacher/Magot-v1-Gemma2-8k-9B-i1-GGUF/resolve/main/Magot-v1-Gemma2-8k-9B.i1-Q3_K_L.gguf) | i1-Q3_K_L | 5.2 | IQ3_M probably better |
49
+ | [GGUF](https://huggingface.co/mradermacher/Magot-v1-Gemma2-8k-9B-i1-GGUF/resolve/main/Magot-v1-Gemma2-8k-9B.i1-IQ4_XS.gguf) | i1-IQ4_XS | 5.3 | |
50
+ | [GGUF](https://huggingface.co/mradermacher/Magot-v1-Gemma2-8k-9B-i1-GGUF/resolve/main/Magot-v1-Gemma2-8k-9B.i1-Q4_0_4_4.gguf) | i1-Q4_0_4_4 | 5.5 | fast on arm, low quality |
51
+ | [GGUF](https://huggingface.co/mradermacher/Magot-v1-Gemma2-8k-9B-i1-GGUF/resolve/main/Magot-v1-Gemma2-8k-9B.i1-Q4_0_4_8.gguf) | i1-Q4_0_4_8 | 5.5 | fast on arm+i8mm, low quality |
52
+ | [GGUF](https://huggingface.co/mradermacher/Magot-v1-Gemma2-8k-9B-i1-GGUF/resolve/main/Magot-v1-Gemma2-8k-9B.i1-Q4_0_8_8.gguf) | i1-Q4_0_8_8 | 5.5 | fast on arm+sve, low quality |
53
+ | [GGUF](https://huggingface.co/mradermacher/Magot-v1-Gemma2-8k-9B-i1-GGUF/resolve/main/Magot-v1-Gemma2-8k-9B.i1-Q4_0.gguf) | i1-Q4_0 | 5.6 | fast, low quality |
54
+ | [GGUF](https://huggingface.co/mradermacher/Magot-v1-Gemma2-8k-9B-i1-GGUF/resolve/main/Magot-v1-Gemma2-8k-9B.i1-Q4_K_S.gguf) | i1-Q4_K_S | 5.6 | optimal size/speed/quality |
55
+ | [GGUF](https://huggingface.co/mradermacher/Magot-v1-Gemma2-8k-9B-i1-GGUF/resolve/main/Magot-v1-Gemma2-8k-9B.i1-Q4_K_M.gguf) | i1-Q4_K_M | 5.9 | fast, recommended |
56
+ | [GGUF](https://huggingface.co/mradermacher/Magot-v1-Gemma2-8k-9B-i1-GGUF/resolve/main/Magot-v1-Gemma2-8k-9B.i1-Q5_K_S.gguf) | i1-Q5_K_S | 6.6 | |
57
+ | [GGUF](https://huggingface.co/mradermacher/Magot-v1-Gemma2-8k-9B-i1-GGUF/resolve/main/Magot-v1-Gemma2-8k-9B.i1-Q5_K_M.gguf) | i1-Q5_K_M | 6.7 | |
58
+ | [GGUF](https://huggingface.co/mradermacher/Magot-v1-Gemma2-8k-9B-i1-GGUF/resolve/main/Magot-v1-Gemma2-8k-9B.i1-Q6_K.gguf) | i1-Q6_K | 7.7 | practically like static Q6_K |
59
+
60
+ Here is a handy graph by ikawrakow comparing some lower-quality quant
61
+ types (lower is better):
62
+
63
+ ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png)
64
+
65
+ And here are Artefact2's thoughts on the matter:
66
+ https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
67
+
68
+ ## FAQ / Model Request
69
+
70
+ See https://huggingface.co/mradermacher/model_requests for some answers to
71
+ questions you might have and/or if you want some other model quantized.
72
+
73
+ ## Thanks
74
+
75
+ I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
76
+ me use its servers and providing upgrades to my workstation to enable
77
+ 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.
78
+
79
+ <!-- end -->