bartowski commited on
Commit
fa035f0
·
verified ·
1 Parent(s): b6c4e18

measurement.json

Browse files
Files changed (2) hide show
  1. README.md +68 -0
  2. measurement.json +0 -0
README.md ADDED
@@ -0,0 +1,68 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ library_name: transformers
4
+ quantized_by: bartowski
5
+ pipeline_tag: text-generation
6
+ ---
7
+
8
+ ## Exllama v2 Quantizations of Marco-o1
9
+
10
+ Using <a href="https://github.com/turboderp/exllamav2/releases/tag/v0.2.4">turboderp's ExLlamaV2 v0.2.4</a> for quantization.
11
+
12
+ <b>The "main" branch only contains the measurement.json, download one of the other branches for the model (see below)</b>
13
+
14
+ Each branch contains an individual bits per weight, with the main one containing only the meaurement.json for further conversions.
15
+
16
+ Original model: https://huggingface.co/AIDC-AI/Marco-o1
17
+
18
+ ## Prompt format
19
+
20
+ ```
21
+ <|im_start|>system
22
+ {system_prompt}<|im_end|>
23
+ <|im_start|>user
24
+ {prompt}<|im_end|>
25
+ <|im_start|>assistant
26
+
27
+ ```
28
+
29
+ ## Available sizes
30
+
31
+
32
+ | Branch | Bits | lm_head bits | VRAM (4k) | VRAM (16k) | VRAM (32k) | Description |
33
+ | ----- | ---- | ------- | ------ | ------ | ------ | ------------ |
34
+ | [8_0](https://huggingface.co/bartowski/Marco-o1-exl2/tree/8_0) | 8.0 | 8.0 | 8.4 GB | 9.8 GB | 11.8 GB | Maximum quality that ExLlamaV2 can produce, near unquantized performance. |
35
+ | [6_5](https://huggingface.co/bartowski/Marco-o1-exl2/tree/6_5) | 6.5 | 8.0 | 7.2 GB | 8.6 GB | 10.6 GB | Very similar to 8.0, good tradeoff of size vs performance, **recommended**. |
36
+ | [5_0](https://huggingface.co/bartowski/Marco-o1-exl2/tree/5_0) | 5.0 | 6.0 | 6.0 GB | 7.4 GB | 9.4 GB | Slightly lower quality vs 6.5, but usable on 8GB cards. |
37
+ | [4_25](https://huggingface.co/bartowski/Marco-o1-exl2/tree/4_25) | 4.25 | 6.0 | 5.3 GB | 6.7 GB | 8.7 GB | GPTQ equivalent bits per weight, slightly higher quality. |
38
+ | [3_5](https://huggingface.co/bartowski/Marco-o1-exl2/tree/3_5) | 3.5 | 6.0 | 4.7 GB | 6.1 GB | 8.1 GB | Lower quality, only use if you have to. |
39
+
40
+ ## Download instructions
41
+
42
+ With git:
43
+
44
+ ```shell
45
+ git clone --single-branch --branch 6_5 https://huggingface.co/bartowski/Marco-o1-exl2 Marco-o1-exl2-6_5
46
+ ```
47
+
48
+ With huggingface hub (credit to TheBloke for instructions):
49
+
50
+ ```shell
51
+ pip3 install huggingface-hub
52
+ ```
53
+
54
+ To download a specific branch, use the `--revision` parameter. For example, to download the 6.5 bpw branch:
55
+
56
+ Linux:
57
+
58
+ ```shell
59
+ huggingface-cli download bartowski/Marco-o1-exl2 --revision 6_5 --local-dir Marco-o1-exl2-6_5
60
+ ```
61
+
62
+ Windows (which apparently doesn't like _ in folders sometimes?):
63
+
64
+ ```shell
65
+ huggingface-cli download bartowski/Marco-o1-exl2 --revision 6_5 --local-dir Marco-o1-exl2-6.5
66
+ ```
67
+
68
+ Want to support my work? Visit my ko-fi page here: https://ko-fi.com/bartowski
measurement.json ADDED
The diff for this file is too large to render. See raw diff