Commit
•
9085a5d
1
Parent(s):
67209cc
"Update README.md"
Browse files
README.md
CHANGED
@@ -8,23 +8,26 @@ license: apache-2.0
|
|
8 |
---
|
9 |
[![banner](https://maddes8cht.github.io/assets/buttons/Huggingface-banner.jpg)]()
|
10 |
|
11 |
-
I
|
12 |
|
13 |
# falcon-40b-instruct - GGUF
|
14 |
- Model creator: [tiiuae](https://huggingface.co/tiiuae)
|
15 |
- Original model: [falcon-40b-instruct](https://huggingface.co/tiiuae/falcon-40b-instruct)
|
16 |
|
17 |
-
#
|
18 |
|
19 |
-
As noted on the [Llama.cpp
|
20 |
|
21 |
-
|
22 |
|
23 |
-
**
|
24 |
-
|
25 |
-
**
|
26 |
-
|
27 |
-
|
|
|
|
|
|
|
28 |
|
29 |
|
30 |
|
@@ -35,6 +38,8 @@ Tiiuae-Falcon 40B instruct is the original instruction following Falcon model fr
|
|
35 |
Falcon is a foundational large language model coming in different sizes: 7b, 40b and 180b.
|
36 |
Sadly, as the Falcon 180b Models are note really free models, I do not provide quantized versions here.
|
37 |
|
|
|
|
|
38 |
|
39 |
|
40 |
# About GGUF format
|
|
|
8 |
---
|
9 |
[![banner](https://maddes8cht.github.io/assets/buttons/Huggingface-banner.jpg)]()
|
10 |
|
11 |
+
I'm constantly enhancing these model descriptions to provide you with the most relevant and comprehensive information
|
12 |
|
13 |
# falcon-40b-instruct - GGUF
|
14 |
- Model creator: [tiiuae](https://huggingface.co/tiiuae)
|
15 |
- Original model: [falcon-40b-instruct](https://huggingface.co/tiiuae/falcon-40b-instruct)
|
16 |
|
17 |
+
# Important Update for Falcon Models in llama.cpp Versions After October 18, 2023
|
18 |
|
19 |
+
As noted on the [Llama.cpp GitHub repository](https://github.com/ggerganov/llama.cpp#hot-topics), all new Llama.cpp releases after October 18, 2023, will require a re-quantization due to the new BPE tokenizer.
|
20 |
|
21 |
+
**Good news!** I am glad that my re-quantization process for Falcon Models is nearly complete. Download the latest quantized models to ensure compatibility with recent llama.cpp software.
|
22 |
|
23 |
+
**Key Points:**
|
24 |
+
|
25 |
+
- **Stay Informed:** Keep an eye on software application release schedules using llama.cpp libraries.
|
26 |
+
- **Monitor Upload Times:** Re-quantization is *almost* done. Watch for updates on my Hugging Face Model pages.
|
27 |
+
|
28 |
+
**Important Compatibility Note:** Old software will work with old Falcon models, but expect updated software to exclusively support the new models.
|
29 |
+
|
30 |
+
This change primarily affects **Falcon** and **Starcoder** models, with other models remaining unaffected.
|
31 |
|
32 |
|
33 |
|
|
|
38 |
Falcon is a foundational large language model coming in different sizes: 7b, 40b and 180b.
|
39 |
Sadly, as the Falcon 180b Models are note really free models, I do not provide quantized versions here.
|
40 |
|
41 |
+
---
|
42 |
+
|
43 |
|
44 |
|
45 |
# About GGUF format
|