update readme.md
Browse files
readme.md
CHANGED
@@ -1,11 +1,34 @@
|
|
1 |
-
|
2 |
-
|
3 |
-
|
4 |
-
|
5 |
-
|
6 |
-
|
7 |
-
|
8 |
-
|
9 |
-
|
10 |
-
|
11 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
license: apache-2.0
|
2 |
+
pipeline_tag: text-to-speech
|
3 |
+
language:
|
4 |
+
- en
|
5 |
+
tags:
|
6 |
+
- model_hub_mixin
|
7 |
+
- pytorch_model_hub_mixin
|
8 |
+
widget:
|
9 |
+
- text: >-
|
10 |
+
# Quantized Dia 1.6B (INT8)
|
11 |
+
|
12 |
+
This is a dynamic int8 quantized version of [nari-labs/Dia-1.6B](https://huggingface.co/nari-labs/Dia-1.6B).
|
13 |
+
It uses dynamic quantization for lighter deployment and faster inference.
|
14 |
+
|
15 |
+
Original model: **float16**, ~6.4GB
|
16 |
+
Quantized model: **int8 dynamic**, ~6.4GB
|
17 |
+
|
18 |
+
Uploaded by [RobertAgee](https://github.com/RobertAgee) and [RobAgrees](https://huggingface.co/RobAgrees.
|
19 |
+
|
20 |
+
> Quantized automatically with PyTorch dynamic quantization in Google Colab.
|
21 |
+
|
22 |
+
- text: >-
|
23 |
+
[S1] Dia is an open weights text to dialogue model. [S2] You get full
|
24 |
+
control over scripts and voices. [S1] Wow. Amazing. (laughs) [S2] Try it
|
25 |
+
now on Git hub or Hugging Face.
|
26 |
+
example_title: Dia intro
|
27 |
+
- text: >-
|
28 |
+
[S1] Oh fire! Oh my goodness! What's the procedure? What to we do people?
|
29 |
+
The smoke could be coming through an air duct! [S2] Oh my god! Okay.. it's
|
30 |
+
happening. Everybody stay calm! [S1] What's the procedure... [S2]
|
31 |
+
Everybody stay fucking calm!!!... Everybody fucking calm down!!!!! [S1]
|
32 |
+
No! No! If you touch the handle, if its hot there might be a fire down the
|
33 |
+
hallway!
|
34 |
+
example_title: Panic protocol
|