Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language:
|
| 3 |
+
- zh
|
| 4 |
+
base_model:
|
| 5 |
+
- Seikaijyu/RWKV6-3B-v2.1-Aphrodite-yandere-chat
|
| 6 |
+
tags:
|
| 7 |
+
- quantization
|
| 8 |
+
quantized_by: btaskel
|
| 9 |
+
---
|
| 10 |
+
From Seikaijyu/RWKV6-3B-v2.1-Aphrodite-yandere-chat:
|
| 11 |
+
https://huggingface.co/Seikaijyu/RWKV6-3B-v2.1-Aphrodite-yandere-chat
|
| 12 |
+
|
| 13 |
+
Based on my experience, Q4_K_S and Q4_K_S are usually the balance points between model size, quantization, and speed.
|
| 14 |
+
|
| 15 |
+
In some benchmarks, selecting a large-parameter low-quantization LLM tends to perform better than a small-parameter high-quantization LLM.
|
| 16 |
+
|
| 17 |
+
根据我的经验，通常Q4_K_S、Q4_K_S是模型尺寸/量化/速度的平衡点
|
| 18 |
+
在一些基准测试下，选择大参数低量化的LLM，要比小参数高量化的LLM要好
|