mradermacher commited on
Commit
50d6309
·
verified ·
1 Parent(s): 2720ef3

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +43 -0
README.md CHANGED
@@ -1 +1,44 @@
 
 
 
 
 
 
 
 
 
1
  weighted/imatrix quants of https://huggingface.co/sophosympatheia/Midnight-Rose-70B-v2.0.3
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - en
4
+ library_name: transformers
5
+ license: llama2
6
+ quantized_by: mradermacher
7
+ ---
8
+ ## About
9
+
10
  weighted/imatrix quants of https://huggingface.co/sophosympatheia/Midnight-Rose-70B-v2.0.3
11
+
12
+
13
+ <!-- provided-files -->
14
+ static quants are available at https://huggingface.co/mradermacher/Midnight-Rose-70B-v2.0.3-GGUF
15
+ ## Usage
16
+
17
+ If you are unsure how to use GGUF files, refer to one of [TheBloke's
18
+ READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
19
+ more details, including on how to concatenate multi-part files.
20
+
21
+ ## Provided Quants
22
+
23
+ (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
24
+
25
+ | Link | Type | Size/GB | Notes |
26
+ |:-----|:-----|--------:|:------|
27
+ | [GGUF](https://huggingface.co/mradermacher/Midnight-Rose-70B-v2.0.3-i1-GGUF/resolve/main/Midnight-Rose-70B-v2.0.3.i1-Q2_K.gguf) | i1-Q2_K | 25.9 | IQ3_XXS probably better |
28
+
29
+
30
+ Here is a handy graph by ikawrakow comparing some lower-quality quant
31
+ types (lower is better):
32
+
33
+ ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png)
34
+
35
+ And here are Artefact2's thoughts on the matter:
36
+ https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
37
+
38
+ ## Thanks
39
+
40
+ I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
41
+ me use its servers and providing upgrades to my workstation to enable
42
+ this work in my free time.
43
+
44
+ <!-- end -->