mradermacher commited on
Commit
3ae7810
·
verified ·
1 Parent(s): 8b4c7db

auto-patch README.md

Browse files
Files changed (1) hide show
  1. README.md +64 -0
README.md CHANGED
@@ -1,6 +1,70 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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/Snowflake/snowflake-arctic-base
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: Snowflake/snowflake-arctic-base
3
+ language:
4
+ - en
5
+ library_name: transformers
6
+ license: apache-2.0
7
+ quantized_by: mradermacher
8
+ tags:
9
+ - snowflake
10
+ - arctic
11
+ - moe
12
+ ---
13
+ ## About
14
+
15
  <!-- ### quantize_version: 2 -->
16
  <!-- ### output_tensor_quantised: 1 -->
17
  <!-- ### convert_type: hf -->
18
  <!-- ### vocab_type: -->
19
  <!-- ### tags: nicoboss -->
20
  weighted/imatrix quants of https://huggingface.co/Snowflake/snowflake-arctic-base
21
+
22
+ <!-- provided-files -->
23
+ static quants are available at https://huggingface.co/mradermacher/snowflake-arctic-base-GGUF
24
+ ## Usage
25
+
26
+ If you are unsure how to use GGUF files, refer to one of [TheBloke's
27
+ READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
28
+ more details, including on how to concatenate multi-part files.
29
+
30
+ ## Provided Quants
31
+
32
+ (sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
33
+
34
+ | Link | Type | Size/GB | Notes |
35
+ |:-----|:-----|--------:|:------|
36
+ | [PART 1](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-IQ1_M.gguf.part1of3) [PART 2](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-IQ1_M.gguf.part2of3) [PART 3](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-IQ1_M.gguf.part3of3) | i1-IQ1_M | 107.4 | mostly desperate |
37
+ | [PART 1](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-IQ2_XXS.gguf.part1of3) [PART 2](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-IQ2_XXS.gguf.part2of3) [PART 3](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-IQ2_XXS.gguf.part3of3) | i1-IQ2_XXS | 125.7 | |
38
+ | [PART 1](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-IQ2_M.gguf.part1of4) [PART 2](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-IQ2_M.gguf.part2of4) [PART 3](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-IQ2_M.gguf.part3of4) [PART 4](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-IQ2_M.gguf.part4of4) | i1-IQ2_M | 156.2 | |
39
+ | [PART 1](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q2_K_S.gguf.part1of4) [PART 2](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q2_K_S.gguf.part2of4) [PART 3](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q2_K_S.gguf.part3of4) [PART 4](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q2_K_S.gguf.part4of4) | i1-Q2_K_S | 162.6 | very low quality |
40
+ | [PART 1](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q2_K.gguf.part1of4) [PART 2](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q2_K.gguf.part2of4) [PART 3](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q2_K.gguf.part3of4) [PART 4](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q2_K.gguf.part4of4) | i1-Q2_K | 174.8 | IQ3_XXS probably better |
41
+ | [PART 1](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-IQ3_XXS.gguf.part1of4) [PART 2](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-IQ3_XXS.gguf.part2of4) [PART 3](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-IQ3_XXS.gguf.part3of4) [PART 4](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-IQ3_XXS.gguf.part4of4) | i1-IQ3_XXS | 184.5 | lower quality |
42
+ | [P1](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q3_K_S.gguf.part1of5) [P2](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q3_K_S.gguf.part2of5) [P3](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q3_K_S.gguf.part3of5) [P4](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q3_K_S.gguf.part4of5) [P5](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q3_K_S.gguf.part5of5) | i1-Q3_K_S | 207.3 | IQ3_XS probably better |
43
+ | [P1](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-IQ3_M.gguf.part1of5) [P2](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-IQ3_M.gguf.part2of5) [P3](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-IQ3_M.gguf.part3of5) [P4](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-IQ3_M.gguf.part4of5) [P5](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-IQ3_M.gguf.part5of5) | i1-IQ3_M | 210.1 | |
44
+ | [P1](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q3_K_M.gguf.part1of5) [P2](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q3_K_M.gguf.part2of5) [P3](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q3_K_M.gguf.part3of5) [P4](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q3_K_M.gguf.part4of5) [P5](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q3_K_M.gguf.part5of5) | i1-Q3_K_M | 229.9 | IQ3_S probably better |
45
+ | [P1](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q3_K_L.gguf.part1of6) [P2](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q3_K_L.gguf.part2of6) [P3](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q3_K_L.gguf.part3of6) [P4](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q3_K_L.gguf.part4of6) [P5](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q3_K_L.gguf.part5of6) [P6](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q3_K_L.gguf.part6of6) | i1-Q3_K_L | 248.9 | IQ3_M probably better |
46
+ | [P1](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-IQ4_XS.gguf.part1of6) [P2](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-IQ4_XS.gguf.part2of6) [P3](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-IQ4_XS.gguf.part3of6) [P4](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-IQ4_XS.gguf.part4of6) [P5](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-IQ4_XS.gguf.part5of6) [P6](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-IQ4_XS.gguf.part6of6) | i1-IQ4_XS | 256.2 | |
47
+ | [P1](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q4_K_S.gguf.part1of6) [P2](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q4_K_S.gguf.part2of6) [P3](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q4_K_S.gguf.part3of6) [P4](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q4_K_S.gguf.part4of6) [P5](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q4_K_S.gguf.part5of6) [P6](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q4_K_S.gguf.part6of6) | i1-Q4_K_S | 273.5 | optimal size/speed/quality |
48
+ | [P1](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q4_K_M.gguf.part1of6) [P2](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q4_K_M.gguf.part2of6) [P3](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q4_K_M.gguf.part3of6) [P4](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q4_K_M.gguf.part4of6) [P5](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q4_K_M.gguf.part5of6) [P6](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q4_K_M.gguf.part6of6) | i1-Q4_K_M | 291.2 | fast, recommended |
49
+ | [P1](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q6_K.gguf.part1of9) [P2](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q6_K.gguf.part2of9) [P3](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q6_K.gguf.part3of9) [P4](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q6_K.gguf.part4of9) [P5](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q6_K.gguf.part5of9) [P6](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q6_K.gguf.part6of9) [P7](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q6_K.gguf.part7of9) [P8](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q6_K.gguf.part8of9) [P9](https://huggingface.co/mradermacher/snowflake-arctic-base-i1-GGUF/resolve/main/snowflake-arctic-base.i1-Q6_K.gguf.part9of9) | i1-Q6_K | 395.3 | practically like static Q6_K |
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
+ ## FAQ / Model Request
60
+
61
+ See https://huggingface.co/mradermacher/model_requests for some answers to
62
+ questions you might have and/or if you want some other model quantized.
63
+
64
+ ## Thanks
65
+
66
+ I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
67
+ me use its servers and providing upgrades to my workstation to enable
68
+ 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.
69
+
70
+ <!-- end -->