Update README.md
Browse files
README.md
CHANGED
|
@@ -6,13 +6,17 @@ license_link: https://github.com/MiniMax-AI/MiniMax-M2.1/blob/main/LICENSE
|
|
| 6 |
|
| 7 |
Exllama 3 quant of Minimax M2.1
|
| 8 |
|
| 9 |
-
Currently have
|
| 10 |
|
| 11 |
4.0bpw H6:
|
| 12 |
- KLD: 0.10375661
|
| 13 |
- Perflexity: 7.79166165
|
| 14 |
|
| 15 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 16 |
6.0bpw H6:
|
| 17 |
- KLD: TBU
|
| 18 |
- Perflexity: TBU
|
|
|
|
| 6 |
|
| 7 |
Exllama 3 quant of Minimax M2.1
|
| 8 |
|
| 9 |
+
Currently have 3 versions:
|
| 10 |
|
| 11 |
4.0bpw H6:
|
| 12 |
- KLD: 0.10375661
|
| 13 |
- Perflexity: 7.79166165
|
| 14 |
|
| 15 |
|
| 16 |
+
4.04bpw H6 (optimized quant):
|
| 17 |
+
- KLD: 0.09264404
|
| 18 |
+
- Perflexity: 7.79524456
|
| 19 |
+
|
| 20 |
6.0bpw H6:
|
| 21 |
- KLD: TBU
|
| 22 |
- Perflexity: TBU
|