Update README.md
Browse files
README.md
CHANGED
|
@@ -6,5 +6,7 @@ tags:
|
|
| 6 |
- yi
|
| 7 |
- 34b
|
| 8 |
---
|
|
|
|
| 9 |
You can run it on 11G mem GPU,quantize base QuIP# method, a weights-only quantization method that is able to achieve near fp16 performance using only 2 bits per weight.
|
|
|
|
| 10 |
url:https://github.com/Cornell-RelaxML/quip-sharp
|
|
|
|
| 6 |
- yi
|
| 7 |
- 34b
|
| 8 |
---
|
| 9 |
+
|
| 10 |
You can run it on 11G mem GPU,quantize base QuIP# method, a weights-only quantization method that is able to achieve near fp16 performance using only 2 bits per weight.
|
| 11 |
+
|
| 12 |
url:https://github.com/Cornell-RelaxML/quip-sharp
|