File size: 4,218 Bytes
d7218ff
 
 
 
 
 
b87b595
d7218ff
 
 
 
 
 
 
 
 
41dbfad
 
d7218ff
 
 
 
 
 
 
 
 
 
 
 
41dbfad
c0547cb
 
a72e6c4
d7de725
9e92ba5
d7de725
 
9e92ba5
d7de725
 
1005f38
9e92ba5
 
1005f38
9e92ba5
 
d7218ff
 
 
 
 
 
 
3b9f4be
 
 
d7218ff
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
---
base_model:
- 152334H/miqu-1-70b-sf
language:
- en
library_name: transformers
license: other
quantized_by: mradermacher
tags:
- mergekit
- merge
---
## About

weighted/imatrix quants of https://huggingface.co/wolfram/miqu-1-103b

<!-- provided-files -->
static quants are available at https://huggingface.co/mradermacher/miqu-1-103b-GGUF
## Usage

If you are unsure how to use GGUF files, refer to one of [TheBloke's
READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
more details, including on how to concatenate multi-part files.

## Provided Quants

(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)

| Link | Type | Size/GB | Notes |
|:-----|:-----|--------:|:------|
| [GGUF](https://huggingface.co/mradermacher/miqu-1-103b-i1-GGUF/resolve/main/miqu-1-103b.i1-IQ1_S.gguf) | i1-IQ1_S | 22.1 | for the desperate |
| [GGUF](https://huggingface.co/mradermacher/miqu-1-103b-i1-GGUF/resolve/main/miqu-1-103b.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 27.7 |  |
| [GGUF](https://huggingface.co/mradermacher/miqu-1-103b-i1-GGUF/resolve/main/miqu-1-103b.i1-IQ2_XS.gguf) | i1-IQ2_XS | 30.8 |  |
| [GGUF](https://huggingface.co/mradermacher/miqu-1-103b-i1-GGUF/resolve/main/miqu-1-103b.i1-IQ2_S.gguf) | i1-IQ2_S | 32.3 |  |
| [GGUF](https://huggingface.co/mradermacher/miqu-1-103b-i1-GGUF/resolve/main/miqu-1-103b.i1-IQ2_M.gguf) | i1-IQ2_M | 35.1 |  |
| [GGUF](https://huggingface.co/mradermacher/miqu-1-103b-i1-GGUF/resolve/main/miqu-1-103b.i1-Q2_K.gguf) | i1-Q2_K | 38.3 | IQ3_XXS probably better |
| [GGUF](https://huggingface.co/mradermacher/miqu-1-103b-i1-GGUF/resolve/main/miqu-1-103b.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 40.0 | fast, lower quality |
| [GGUF](https://huggingface.co/mradermacher/miqu-1-103b-i1-GGUF/resolve/main/miqu-1-103b.i1-IQ3_XS.gguf) | i1-IQ3_XS | 42.5 |  |
| [GGUF](https://huggingface.co/mradermacher/miqu-1-103b-i1-GGUF/resolve/main/miqu-1-103b.i1-Q3_K_S.gguf) | i1-Q3_K_S | 44.9 | IQ3_XS probably better |
| [GGUF](https://huggingface.co/mradermacher/miqu-1-103b-i1-GGUF/resolve/main/miqu-1-103b.i1-IQ3_S.gguf) | i1-IQ3_S | 45.0 | fast, beats Q3_K* |
| [GGUF](https://huggingface.co/mradermacher/miqu-1-103b-i1-GGUF/resolve/main/miqu-1-103b.i1-IQ3_M.gguf) | i1-IQ3_M | 46.5 |  |
| [PART 1](https://huggingface.co/mradermacher/miqu-1-103b-i1-GGUF/resolve/main/miqu-1-103b.i1-Q3_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/miqu-1-103b-i1-GGUF/resolve/main/miqu-1-103b.i1-Q3_K_M.gguf.part2of2) | i1-Q3_K_M | 50.0 | IQ3_S probably better |
| [PART 1](https://huggingface.co/mradermacher/miqu-1-103b-i1-GGUF/resolve/main/miqu-1-103b.i1-Q3_K_L.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/miqu-1-103b-i1-GGUF/resolve/main/miqu-1-103b.i1-Q3_K_L.gguf.part2of2) | i1-Q3_K_L | 54.5 | IQ3_M probably better |
| [PART 1](https://huggingface.co/mradermacher/miqu-1-103b-i1-GGUF/resolve/main/miqu-1-103b.i1-Q4_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/miqu-1-103b-i1-GGUF/resolve/main/miqu-1-103b.i1-Q4_K_S.gguf.part2of2) | i1-Q4_K_S | 59.0 | almost as good as Q4_K_M |
| [PART 1](https://huggingface.co/mradermacher/miqu-1-103b-i1-GGUF/resolve/main/miqu-1-103b.i1-Q4_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/miqu-1-103b-i1-GGUF/resolve/main/miqu-1-103b.i1-Q4_K_M.gguf.part2of2) | i1-Q4_K_M | 62.3 | fast, medium quality |
| [PART 1](https://huggingface.co/mradermacher/miqu-1-103b-i1-GGUF/resolve/main/miqu-1-103b.i1-Q5_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/miqu-1-103b-i1-GGUF/resolve/main/miqu-1-103b.i1-Q5_K_S.gguf.part2of2) | i1-Q5_K_S | 71.4 |  |
| [PART 1](https://huggingface.co/mradermacher/miqu-1-103b-i1-GGUF/resolve/main/miqu-1-103b.i1-Q5_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/miqu-1-103b-i1-GGUF/resolve/main/miqu-1-103b.i1-Q5_K_M.gguf.part2of2) | i1-Q5_K_M | 73.3 | best weighted quant |


Here is a handy graph by ikawrakow comparing some lower-quality quant
types (lower is better):

![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png)

And here are Artefact2's thoughts on the matter:
https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9

<!-- end -->