mradermacher commited on
Commit
23cec28
1 Parent(s): dfdb149

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +68 -0
README.md CHANGED
@@ -1,5 +1,73 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  <!-- ### quantize_version: 2 -->
2
  <!-- ### output_tensor_quantised: 1 -->
3
  <!-- ### convert_type: hf -->
4
  <!-- ### vocab_type: -->
5
  static quants of https://huggingface.co/frost19k/dolphin-2.8-mistral-v02-2x7b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: frost19k/dolphin-2.8-mistral-v02-2x7b
3
+ language:
4
+ - en
5
+ library_name: transformers
6
+ license: apache-2.0
7
+ quantized_by: mradermacher
8
+ tags:
9
+ - moe
10
+ - frankenmoe
11
+ - merge
12
+ - mergekit
13
+ - lazymergekit
14
+ - cognitivecomputations/dolphin-2.8-mistral-7b-v02
15
+ - OpenPipe/mistral-ft-optimized-1227
16
+ ---
17
+ ## About
18
+
19
  <!-- ### quantize_version: 2 -->
20
  <!-- ### output_tensor_quantised: 1 -->
21
  <!-- ### convert_type: hf -->
22
  <!-- ### vocab_type: -->
23
  static quants of https://huggingface.co/frost19k/dolphin-2.8-mistral-v02-2x7b
24
+
25
+ <!-- provided-files -->
26
+ 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.
27
+ ## Usage
28
+
29
+ If you are unsure how to use GGUF files, refer to one of [TheBloke's
30
+ READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
31
+ more details, including on how to concatenate multi-part files.
32
+
33
+ ## Provided Quants
34
+
35
+ (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
36
+
37
+ | Link | Type | Size/GB | Notes |
38
+ |:-----|:-----|--------:|:------|
39
+ | [GGUF](https://huggingface.co/mradermacher/dolphin-2.8-mistral-v02-2x7b-GGUF/resolve/main/dolphin-2.8-mistral-v02-2x7b.Q2_K.gguf) | Q2_K | 4.9 | |
40
+ | [GGUF](https://huggingface.co/mradermacher/dolphin-2.8-mistral-v02-2x7b-GGUF/resolve/main/dolphin-2.8-mistral-v02-2x7b.IQ3_XS.gguf) | IQ3_XS | 5.4 | |
41
+ | [GGUF](https://huggingface.co/mradermacher/dolphin-2.8-mistral-v02-2x7b-GGUF/resolve/main/dolphin-2.8-mistral-v02-2x7b.Q3_K_S.gguf) | Q3_K_S | 5.7 | |
42
+ | [GGUF](https://huggingface.co/mradermacher/dolphin-2.8-mistral-v02-2x7b-GGUF/resolve/main/dolphin-2.8-mistral-v02-2x7b.IQ3_S.gguf) | IQ3_S | 5.7 | beats Q3_K* |
43
+ | [GGUF](https://huggingface.co/mradermacher/dolphin-2.8-mistral-v02-2x7b-GGUF/resolve/main/dolphin-2.8-mistral-v02-2x7b.IQ3_M.gguf) | IQ3_M | 5.8 | |
44
+ | [GGUF](https://huggingface.co/mradermacher/dolphin-2.8-mistral-v02-2x7b-GGUF/resolve/main/dolphin-2.8-mistral-v02-2x7b.Q3_K_M.gguf) | Q3_K_M | 6.3 | lower quality |
45
+ | [GGUF](https://huggingface.co/mradermacher/dolphin-2.8-mistral-v02-2x7b-GGUF/resolve/main/dolphin-2.8-mistral-v02-2x7b.Q3_K_L.gguf) | Q3_K_L | 6.8 | |
46
+ | [GGUF](https://huggingface.co/mradermacher/dolphin-2.8-mistral-v02-2x7b-GGUF/resolve/main/dolphin-2.8-mistral-v02-2x7b.IQ4_XS.gguf) | IQ4_XS | 7.1 | |
47
+ | [GGUF](https://huggingface.co/mradermacher/dolphin-2.8-mistral-v02-2x7b-GGUF/resolve/main/dolphin-2.8-mistral-v02-2x7b.Q4_K_S.gguf) | Q4_K_S | 7.4 | fast, recommended |
48
+ | [GGUF](https://huggingface.co/mradermacher/dolphin-2.8-mistral-v02-2x7b-GGUF/resolve/main/dolphin-2.8-mistral-v02-2x7b.Q4_K_M.gguf) | Q4_K_M | 7.9 | fast, recommended |
49
+ | [GGUF](https://huggingface.co/mradermacher/dolphin-2.8-mistral-v02-2x7b-GGUF/resolve/main/dolphin-2.8-mistral-v02-2x7b.Q5_K_S.gguf) | Q5_K_S | 9.0 | |
50
+ | [GGUF](https://huggingface.co/mradermacher/dolphin-2.8-mistral-v02-2x7b-GGUF/resolve/main/dolphin-2.8-mistral-v02-2x7b.Q5_K_M.gguf) | Q5_K_M | 9.2 | |
51
+ | [GGUF](https://huggingface.co/mradermacher/dolphin-2.8-mistral-v02-2x7b-GGUF/resolve/main/dolphin-2.8-mistral-v02-2x7b.Q6_K.gguf) | Q6_K | 10.7 | very good quality |
52
+ | [GGUF](https://huggingface.co/mradermacher/dolphin-2.8-mistral-v02-2x7b-GGUF/resolve/main/dolphin-2.8-mistral-v02-2x7b.Q8_0.gguf) | Q8_0 | 13.8 | fast, best quality |
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
+ ## FAQ / Model Request
63
+
64
+ See https://huggingface.co/mradermacher/model_requests for some answers to
65
+ questions you might have and/or if you want some other model quantized.
66
+
67
+ ## Thanks
68
+
69
+ I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
70
+ me use its servers and providing upgrades to my workstation to enable
71
+ this work in my free time.
72
+
73
+ <!-- end -->