Create vllm serve
Browse files- vllm serve +12 -0
vllm serve
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# Load and run the model:
|
| 2 |
+
vllm serve "facebook/musicgen-small"
|
| 3 |
+
|
| 4 |
+
# Call the server using curl:
|
| 5 |
+
curl -X POST "http://localhost:8000/v1/chat/completions" \
|
| 6 |
+
-H "Content-Type: application/json" \
|
| 7 |
+
--data '{
|
| 8 |
+
"model": "facebook/musicgen-small"
|
| 9 |
+
"messages": [
|
| 10 |
+
{"role": "user", "content": "Hello!"}
|
| 11 |
+
]
|
| 12 |
+
}'
|