Update README.md
Browse files
README.md
CHANGED
@@ -1,4 +1,7 @@
|
|
1 |
---
|
2 |
license: apache-2.0
|
3 |
---
|
4 |
-
Various models in GGUF format quantized with a new 2-bit approach. Intended for use with llama.cpp. Requires llama.cpp PR 4773.
|
|
|
|
|
|
|
|
1 |
---
|
2 |
license: apache-2.0
|
3 |
---
|
4 |
+
Various models in GGUF format quantized with a new 2-bit approach. Intended for use with llama.cpp. Requires llama.cpp PR 4773.
|
5 |
+
|
6 |
+
Update: PR 4773 has been merged into `llama.cpp`, but I have added new models that require PR 4856.
|
7 |
+
The new models are those that have around 2.3-2.4 bpw. They have a lower quantization error at the xpense of being ~10% larger.
|