Fix/fix qwen3 deploy (#157)

* merge

* merge

* add Mistral-Small-3.1-24B-Instruct-2503

* modify qwq-32b deploy

* add txgemma model;

* modify model list command

* fix typo

* add some ecs parameters

* add glm4-z1 models

* modify vllm backend

* add qwen3

* fix cli bugs

* fix

* add deeseek r1/Qwen3-235B-A22B

* fix local deploy account bug

* add qwen 3 awq models

* fix serialize_utils bugs

* modify qwen3 deployment

* modify docs

* modify qwen3 engine;add strands client test

* modify engine

* merge

Author: 11zhouxuan

src/emd/models/llms/qwen.py

@@ -539,7 +539,7 @@
 Model.register(
     dict(
         model_id = "Qwen3-0.6B",
-        supported_engines=[vllm_qwen3_engin084],
+        supported_engines=[vllm_qwen3_engin091],
         supported_instances=[
             g5d2xlarge_instance,
             g5d4xlarge_instance,
@@ -573,7 +573,7 @@
 Model.register(
     dict(
         model_id = "Qwen3-1.7B",
-        supported_engines=[vllm_qwen3_engin084],
+        supported_engines=[vllm_qwen3_engin091],
         supported_instances=[
             g5d2xlarge_instance,
             g5d4xlarge_instance,
@@ -608,7 +608,7 @@
 Model.register(
     dict(
         model_id = "Qwen3-4B",
-        supported_engines=[vllm_qwen3_engin084],
+        supported_engines=[vllm_qwen3_engin091],
         supported_instances=[
             g5d2xlarge_instance,
             g5d4xlarge_instance,
@@ -644,7 +644,7 @@
 Model.register(
     dict(
         model_id = "Qwen3-14B-AWQ",
-        supported_engines=[vllm_qwen3_engin084],
+        supported_engines=[vllm_qwen3_engin091],
         supported_instances=[
             g5d2xlarge_instance,
             g5d4xlarge_instance,
@@ -679,7 +679,7 @@
 Model.register(
     dict(
         model_id = "Qwen3-14B",
-        supported_engines=[vllm_qwen3_engin084],
+        supported_engines=[vllm_qwen3_engin091],
         supported_instances=[
             g5d12xlarge_instance,
             g5d24xlarge_instance,
@@ -714,7 +714,7 @@
 # Model.register(
 #     dict(
 #         model_id = "Qwen3-14B-FP8",
-#         supported_engines=[vllm_qwen3_engin084],
+#         supported_engines=[vllm_qwen3_engin091],
 #         supported_instances=[
 #             g5d2xlarge_instance,
 #             g5d4xlarge_instance,
@@ -750,7 +750,7 @@
 Model.register(
     dict(
         model_id = "Qwen3-32B-AWQ",
-        supported_engines=[vllm_qwen3_engin084],
+        supported_engines=[vllm_qwen3_engin091],
         supported_instances=[
             g5d12xlarge_instance,
             g5d24xlarge_instance,
@@ -784,7 +784,7 @@
 Model.register(
     dict(
         model_id = "Qwen3-32B",
-        supported_engines=[vllm_qwen3_engin084],
+        supported_engines=[vllm_qwen3_engin091],
         supported_instances=[
             g5d12xlarge_instance,
             g5d24xlarge_instance,
@@ -817,7 +817,7 @@
 Model.register(
     dict(
         model_id = "Qwen3-30B-A3B",
-        supported_engines=[vllm_qwen3_engin084],
+        supported_engines=[vllm_qwen3_engin091],
         supported_instances=[
             g5d12xlarge_instance,
             g5d24xlarge_instance,
@@ -850,7 +850,7 @@
 Model.register(
     dict(
         model_id = "Qwen3-235B-A22B",
-        supported_engines=[vllm_qwen3_engin084],
+        supported_engines=[vllm_qwen3_engin091],
         supported_instances=[
             local_instance
         ],
@@ -874,7 +874,7 @@
 Model.register(
     dict(
         model_id = "Qwen3-235B-A22B-FP8",
-        supported_engines=[vllm_qwen3_engin084],
+        supported_engines=[vllm_qwen3_engin091],
         supported_instances=[
             local_instance
         ],