[model] Qwen3-30B-A3B-Thinking-2507 (#8788)

setup.py

@@ -52,7 +52,7 @@ extra_require = {
     "eetq": ["eetq"],
     "gptq": ["optimum>=1.24.0", "gptqmodel>=2.0.0"],
     "aqlm": ["aqlm[gpu]>=1.1.0"],
-    "vllm": ["vllm>=0.4.3,<=0.9.1"],
+    "vllm": ["vllm>=0.4.3,<=0.10.0"],
     "sglang": ["sglang[srt]>=0.4.5", "transformers==4.51.1"],
     "galore": ["galore-torch"],
     "apollo": ["apollo-torch"],

src/llamafactory/extras/constants.py

@@ -2747,6 +2747,10 @@ register_model_group(
             DownloadSource.DEFAULT: "Qwen/Qwen3-30B-A3B-Instruct-2507",
             DownloadSource.MODELSCOPE: "Qwen/Qwen3-30B-A3B-Instruct-2507",
         },
+        "Qwen3-30B-A3B-Thinking-2507": {
+            DownloadSource.DEFAULT: "Qwen/Qwen3-30B-A3B-Thinking-2507",
+            DownloadSource.MODELSCOPE: "Qwen/Qwen3-30B-A3B-Thinking-2507",
+        },
         "Qwen3-235B-A22B-Thinking": {
             DownloadSource.DEFAULT: "Qwen/Qwen3-235B-A22B",
             DownloadSource.MODELSCOPE: "Qwen/Qwen3-235B-A22B",

src/llamafactory/hparams/parser.py

@@ -146,7 +146,7 @@ def _check_extra_dependencies(
         check_version("mixture-of-depth>=1.1.6", mandatory=True)
 
     if model_args.infer_backend == EngineName.VLLM:
-        check_version("vllm>=0.4.3,<=0.9.1")
+        check_version("vllm>=0.4.3,<=0.10.0")
         check_version("vllm", mandatory=True)
     elif model_args.infer_backend == EngineName.SGLANG:
         check_version("sglang>=0.4.5")