Update README.md
Browse files
README.md
CHANGED
@@ -4,9 +4,10 @@ tags:
|
|
4 |
- GGUF
|
5 |
---
|
6 |
|
7 |
-
#
|
|
|
8 |
|
9 |
-
|
10 |
|
11 |
# Details
|
12 |
[Thanks to mlabonne for the initial code](https://huggingface.co/mlabonne)
|
@@ -17,6 +18,4 @@ RP Imatrix is from [Lewdiculous](https://huggingface.co/Lewdiculous)
|
|
17 |
|
18 |
Host files for a google colab notebook, hoping to make it easier to GGUF models with Imatrix.
|
19 |
|
20 |
-
There are two imatrix datasets, one for general use and one for RP.
|
21 |
-
|
22 |
-
After some testing, making the actual quants is really slow, recommended to only use it for the intial FP16 GGUF and imatrix.dat generation.
|
|
|
4 |
- GGUF
|
5 |
---
|
6 |
|
7 |
+
# Free Tier Colab
|
8 |
+
This is only for making the intial FP16 gguf file and computing an imatrix.dat
|
9 |
|
10 |
+
Quantizing is too slow on colab due to only having two available cores.
|
11 |
|
12 |
# Details
|
13 |
[Thanks to mlabonne for the initial code](https://huggingface.co/mlabonne)
|
|
|
18 |
|
19 |
Host files for a google colab notebook, hoping to make it easier to GGUF models with Imatrix.
|
20 |
|
21 |
+
There are two imatrix datasets, one for general use and one for RP.
|
|
|
|