remove tensor parallelism restriction

VipulMascarenhas · VipulMascarenhas · commit 6d8873e87b43 · 2024-07-02T14:09:10.000-07:00
diff --git a/ads/aqua/constants.py b/ads/aqua/constants.py
@@ -62,7 +62,6 @@
 }
 
 VLLM_INFERENCE_RESTRICTED_PARAMS = {
-    "--tensor-parallel-size",
     "--port",
     "--host",
     "--served-model-name",
diff --git a/tests/unitary/with_extras/aqua/test_deployment.py b/tests/unitary/with_extras/aqua/test_deployment.py
@@ -492,7 +492,7 @@ def test_get_deployment_default_params(
             ),
             (
                 "odsc-vllm-serving",
-                ["--tensor-parallel-size 2"],
+                ["--port 8081"],
             ),
             (
                 "odsc-tgi-serving",

Original file line number	Diff line number	Diff line change
`@@ -62,7 +62,6 @@`
`62`	`62`	`}`
`63`	`63`
`64`	`64`	`VLLM_INFERENCE_RESTRICTED_PARAMS = {`
`65`		`- "--tensor-parallel-size",`
`66`	`65`	`"--port",`
`67`	`66`	`"--host",`
`68`	`67`	`"--served-model-name",`
Original file line number	Diff line number	Diff line change
`@@ -492,7 +492,7 @@ def test_get_deployment_default_params(`
`492`	`492`	`),`
`493`	`493`	`(`
`494`	`494`	`"odsc-vllm-serving",`
`495`		`- ["--tensor-parallel-size 2"],`
	`495`	`+ ["--port 8081"],`
`496`	`496`	`),`
`497`	`497`	`(`
`498`	`498`	`"odsc-tgi-serving",`