Update vllm_plugin/SERVING.md
Browse files- vllm_plugin/SERVING.md +2 -0
vllm_plugin/SERVING.md
CHANGED
|
@@ -2,6 +2,8 @@
|
|
| 2 |
|
| 3 |
## Prerequisites
|
| 4 |
|
|
|
|
|
|
|
| 5 |
- NVIDIA Blackwell GPU (B300 / B200 / RTX 5090) for real Quartet II NVFP4 kernels
|
| 6 |
- CUDA 13.0+
|
| 7 |
- Python 3.11+
|
|
|
|
| 2 |
|
| 3 |
## Prerequisites
|
| 4 |
|
| 5 |
+
Before following this guide, first set up the environment as described in `lm_eval/README.md`.
|
| 6 |
+
|
| 7 |
- NVIDIA Blackwell GPU (B300 / B200 / RTX 5090) for real Quartet II NVFP4 kernels
|
| 8 |
- CUDA 13.0+
|
| 9 |
- Python 3.11+
|