mradermacher commited on
Commit
38d513c
·
verified ·
1 Parent(s): e097cb6

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +64 -0
README.md CHANGED
@@ -1 +1,65 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  weighted/imatrix quants of https://huggingface.co/DenisTheDev/Blitz-AI-ULTRA
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ exported_from: DenisTheDev/Blitz-AI-ULTRA
3
+ language:
4
+ - en
5
+ library_name: transformers
6
+ license: apache-2.0
7
+ quantized_by: mradermacher
8
+ tags:
9
+ - merge
10
+ - mergekit
11
+ - lazymergekit
12
+ ---
13
+ ## About
14
+
15
  weighted/imatrix quants of https://huggingface.co/DenisTheDev/Blitz-AI-ULTRA
16
+
17
+
18
+ <!-- provided-files -->
19
+ static quants are available at https://huggingface.co/mradermacher/Blitz-AI-ULTRA-GGUF
20
+ ## Usage
21
+
22
+ If you are unsure how to use GGUF files, refer to one of [TheBloke's
23
+ READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
24
+ more details, including on how to concatenate multi-part files.
25
+
26
+ ## Provided Quants
27
+
28
+ (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
29
+
30
+ | Link | Type | Size/GB | Notes |
31
+ |:-----|:-----|--------:|:------|
32
+ | [GGUF](https://huggingface.co/mradermacher/Blitz-AI-ULTRA-i1-GGUF/resolve/main/Blitz-AI-ULTRA.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 38.8 | |
33
+ | [GGUF](https://huggingface.co/mradermacher/Blitz-AI-ULTRA-i1-GGUF/resolve/main/Blitz-AI-ULTRA.i1-IQ2_M.gguf) | i1-IQ2_M | 48.4 | |
34
+ | [PART 1](https://huggingface.co/mradermacher/Blitz-AI-ULTRA-i1-GGUF/resolve/main/Blitz-AI-ULTRA.i1-Q2_K.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Blitz-AI-ULTRA-i1-GGUF/resolve/main/Blitz-AI-ULTRA.i1-Q2_K.gguf.part2of2) | i1-Q2_K | 51.6 | IQ3_XXS probably better |
35
+ | [PART 1](https://huggingface.co/mradermacher/Blitz-AI-ULTRA-i1-GGUF/resolve/main/Blitz-AI-ULTRA.i1-IQ3_XXS.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Blitz-AI-ULTRA-i1-GGUF/resolve/main/Blitz-AI-ULTRA.i1-IQ3_XXS.gguf.part2of2) | i1-IQ3_XXS | 53.0 | lower quality |
36
+ | [PART 1](https://huggingface.co/mradermacher/Blitz-AI-ULTRA-i1-GGUF/resolve/main/Blitz-AI-ULTRA.i1-IQ3_XS.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Blitz-AI-ULTRA-i1-GGUF/resolve/main/Blitz-AI-ULTRA.i1-IQ3_XS.gguf.part2of2) | i1-IQ3_XS | 56.7 | |
37
+ | [PART 1](https://huggingface.co/mradermacher/Blitz-AI-ULTRA-i1-GGUF/resolve/main/Blitz-AI-ULTRA.i1-IQ3_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Blitz-AI-ULTRA-i1-GGUF/resolve/main/Blitz-AI-ULTRA.i1-IQ3_S.gguf.part2of2) | i1-IQ3_S | 59.6 | beats Q3_K* |
38
+ | [PART 1](https://huggingface.co/mradermacher/Blitz-AI-ULTRA-i1-GGUF/resolve/main/Blitz-AI-ULTRA.i1-Q3_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Blitz-AI-ULTRA-i1-GGUF/resolve/main/Blitz-AI-ULTRA.i1-Q3_K_S.gguf.part2of2) | i1-Q3_K_S | 59.6 | IQ3_XS probably better |
39
+ | [PART 1](https://huggingface.co/mradermacher/Blitz-AI-ULTRA-i1-GGUF/resolve/main/Blitz-AI-ULTRA.i1-IQ3_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Blitz-AI-ULTRA-i1-GGUF/resolve/main/Blitz-AI-ULTRA.i1-IQ3_M.gguf.part2of2) | i1-IQ3_M | 62.7 | |
40
+ | [PART 1](https://huggingface.co/mradermacher/Blitz-AI-ULTRA-i1-GGUF/resolve/main/Blitz-AI-ULTRA.i1-Q3_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Blitz-AI-ULTRA-i1-GGUF/resolve/main/Blitz-AI-ULTRA.i1-Q3_K_M.gguf.part2of2) | i1-Q3_K_M | 66.3 | IQ3_S probably better |
41
+ | [PART 1](https://huggingface.co/mradermacher/Blitz-AI-ULTRA-i1-GGUF/resolve/main/Blitz-AI-ULTRA.i1-Q3_K_L.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Blitz-AI-ULTRA-i1-GGUF/resolve/main/Blitz-AI-ULTRA.i1-Q3_K_L.gguf.part2of2) | i1-Q3_K_L | 72.1 | IQ3_M probably better |
42
+ | [PART 1](https://huggingface.co/mradermacher/Blitz-AI-ULTRA-i1-GGUF/resolve/main/Blitz-AI-ULTRA.i1-IQ4_XS.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Blitz-AI-ULTRA-i1-GGUF/resolve/main/Blitz-AI-ULTRA.i1-IQ4_XS.gguf.part2of2) | i1-IQ4_XS | 72.5 | |
43
+ | [PART 1](https://huggingface.co/mradermacher/Blitz-AI-ULTRA-i1-GGUF/resolve/main/Blitz-AI-ULTRA.i1-Q4_0.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Blitz-AI-ULTRA-i1-GGUF/resolve/main/Blitz-AI-ULTRA.i1-Q4_0.gguf.part2of2) | i1-Q4_0 | 76.7 | |
44
+ | [PART 1](https://huggingface.co/mradermacher/Blitz-AI-ULTRA-i1-GGUF/resolve/main/Blitz-AI-ULTRA.i1-Q4_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Blitz-AI-ULTRA-i1-GGUF/resolve/main/Blitz-AI-ULTRA.i1-Q4_K_S.gguf.part2of2) | i1-Q4_K_S | 77.0 | optimal size/speed/quality |
45
+ | [PART 1](https://huggingface.co/mradermacher/Blitz-AI-ULTRA-i1-GGUF/resolve/main/Blitz-AI-ULTRA.i1-Q4_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Blitz-AI-ULTRA-i1-GGUF/resolve/main/Blitz-AI-ULTRA.i1-Q4_K_M.gguf.part2of2) | i1-Q4_K_M | 81.4 | fast, recommended |
46
+ | [PART 1](https://huggingface.co/mradermacher/Blitz-AI-ULTRA-i1-GGUF/resolve/main/Blitz-AI-ULTRA.i1-Q5_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Blitz-AI-ULTRA-i1-GGUF/resolve/main/Blitz-AI-ULTRA.i1-Q5_K_S.gguf.part2of2) | i1-Q5_K_S | 92.3 | |
47
+ | [PART 1](https://huggingface.co/mradermacher/Blitz-AI-ULTRA-i1-GGUF/resolve/main/Blitz-AI-ULTRA.i1-Q5_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Blitz-AI-ULTRA-i1-GGUF/resolve/main/Blitz-AI-ULTRA.i1-Q5_K_M.gguf.part2of2) | i1-Q5_K_M | 94.9 | |
48
+ | [PART 1](https://huggingface.co/mradermacher/Blitz-AI-ULTRA-i1-GGUF/resolve/main/Blitz-AI-ULTRA.i1-Q6_K.gguf.part1of3) [PART 2](https://huggingface.co/mradermacher/Blitz-AI-ULTRA-i1-GGUF/resolve/main/Blitz-AI-ULTRA.i1-Q6_K.gguf.part2of3) [PART 3](https://huggingface.co/mradermacher/Blitz-AI-ULTRA-i1-GGUF/resolve/main/Blitz-AI-ULTRA.i1-Q6_K.gguf.part3of3) | i1-Q6_K | 109.2 | practically like static Q6_K |
49
+
50
+
51
+ Here is a handy graph by ikawrakow comparing some lower-quality quant
52
+ types (lower is better):
53
+
54
+ ![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png)
55
+
56
+ And here are Artefact2's thoughts on the matter:
57
+ https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
58
+
59
+ ## Thanks
60
+
61
+ I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
62
+ me use its servers and providing upgrades to my workstation to enable
63
+ this work in my free time.
64
+
65
+ <!-- end -->