Update README.md
Browse files
README.md
CHANGED
|
@@ -11,6 +11,15 @@ The TPS (tokens per second) improvement reached nearly 70%.
|
|
| 11 |
|
| 12 |
To use Eagle3 with SGLang, first replace the qwen3_moe.py file in SGLang’s directory (sglang/python/sglang/srt/models/) with the qwen3_moe.py file from this project.
|
| 13 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 14 |
The launch command for using Eagle3 with SGLang is:
|
| 15 |
|
| 16 |
```python3
|
|
|
|
| 11 |
|
| 12 |
To use Eagle3 with SGLang, first replace the qwen3_moe.py file in SGLang’s directory (sglang/python/sglang/srt/models/) with the qwen3_moe.py file from this project.
|
| 13 |
|
| 14 |
+
|
| 15 |
+
| model | gpu | tps |
|
| 16 |
+
|---------|---------|---------|
|
| 17 |
+
| qwen3-8b | h200 | 147 |
|
| 18 |
+
| qwen3-8b-eagle3 | h200 | 231 |
|
| 19 |
+
| qwen3-8b | 8*h200 | 183 |
|
| 20 |
+
| qwen3-8b-eagle3 | 8*h200 | 325 |
|
| 21 |
+
|
| 22 |
+
|
| 23 |
The launch command for using Eagle3 with SGLang is:
|
| 24 |
|
| 25 |
```python3
|