Add Qwen3-Omni moe thinker (#25550)

Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Signed-off-by: Roger Wang <hey@rogerw.io>
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Co-authored-by: Xiong Wang <feizi.wx@alibaba-inc.com>
Co-authored-by: DarkLight1337 <tlleungac@connect.ust.hk>
Co-authored-by: Roger Wang <hey@rogerw.io>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>

docs/models/supported_models.md

@@ -714,6 +714,7 @@ These models primarily accept the [`LLM.generate`](./generative_models.md#llmgen
 | `Qwen2_5OmniThinkerForConditionalGeneration` | Qwen2.5-Omni | T + I<sup>E+</sup> + V<sup>E+</sup> + A<sup>+</sup> | `Qwen/Qwen2.5-Omni-3B`, `Qwen/Qwen2.5-Omni-7B` | ✅︎ | ✅︎ | ✅︎ |
 | `Qwen3VLForConditionalGeneration` | Qwen3-VL | T + I<sup>E+</sup> + V<sup>E+</sup> | `Qwen/Qwen3-VL-4B-Instruct`, etc. | ✅︎ | ✅︎ | ✅︎ |
 | `Qwen3VLMoeForConditionalGeneration` | Qwen3-VL-MOE | T + I<sup>E+</sup> + V<sup>E+</sup> | `Qwen/Qwen3-VL-30B-A3B-Instruct`, etc. | ✅︎ | ✅︎ | ✅︎ |
+| `Qwen3OmniMoeThinkerForConditionalGeneration` | Qwen3-Omni | T + I<sup>E+</sup> + V<sup>E+</sup> + A<sup>+</sup> | `Qwen/Qwen3-Omni-30B-A3B-Instruct`, `Qwen/Qwen3-Omni-30B-A3B-Thinking` | ✅︎ | ✅︎ | ✅︎ |
 | `RForConditionalGeneration` | R-VL-4B | T + I<sup>E+</sup> | `YannQi/R-4B` | | ✅︎ | ✅︎ |
 | `SkyworkR1VChatModel` | Skywork-R1V-38B | T + I | `Skywork/Skywork-R1V-38B` | | ✅︎ | ✅︎ |
 | `SmolVLMForConditionalGeneration` | SmolVLM2 | T + I | `SmolVLM2-2.2B-Instruct` | ✅︎ | | ✅︎ |
@@ -804,8 +805,7 @@ Some models are supported only via the [Transformers backend](#transformers). Th
     Our PaliGemma implementations have the same problem as Gemma 3 (see above) for both V0 and V1.
 
 !!! note
-    For Qwen2.5-Omni, reading audio from video pre-processing (`--mm-processor-kwargs '{"use_audio_in_video": true}'`)
-    is currently supported on V0 (but not V1), because overlapping modalities is not yet supported in V1.
+    For Qwen2.5-Omni and Qwen3-Omni, reading audio from video pre-processing (`--mm-processor-kwargs '{"use_audio_in_video": true}'`) is currently work in progress and not yet supported.
 
 #### Transcription
 

tests/models/multimodal/processing/test_common.py

@@ -384,6 +384,7 @@ def _test_processing_correctness_one(
         "Qwen/Qwen2.5-Omni-3B",
         "Qwen/Qwen3-VL-4B-Instruct",
         "Qwen/Qwen3-VL-30B-A3B-Instruct",
+        "Qwen/Qwen3-Omni-30B-A3B-Instruct",
         "YannQi/R-4B",
         "Skywork/Skywork-R1V-38B",
         "HuggingFaceTB/SmolVLM2-2.2B-Instruct",

tests/models/registry.py

@@ -773,6 +773,11 @@ _MULTIMODAL_EXAMPLE_MODELS = {
         min_transformers_version="4.57",
         is_available_online=False,
     ),
+    "Qwen3OmniMoeForConditionalGeneration": _HfExamplesInfo(
+        "Qwen/Qwen3-Omni-30B-A3B-Instruct",
+        max_model_len=4096,
+        min_transformers_version="4.57",
+    ),
     "RForConditionalGeneration": _HfExamplesInfo("YannQi/R-4B", trust_remote_code=True),
     "SkyworkR1VChatModel": _HfExamplesInfo(
         "Skywork/Skywork-R1V-38B", trust_remote_code=True

vllm/model_executor/models/registry.py

@@ -355,6 +355,10 @@ _MULTIMODAL_MODELS = {
         "qwen2_5_omni_thinker",
         "Qwen2_5OmniThinkerForConditionalGeneration",
     ),
+    "Qwen3OmniMoeForConditionalGeneration": (
+        "qwen3_omni_moe_thinker",
+        "Qwen3OmniMoeThinkerForConditionalGeneration",
+    ),
     "Qwen3VLForConditionalGeneration": ("qwen3_vl", "Qwen3VLForConditionalGeneration"),  # noqa: E501
     "Qwen3VLMoeForConditionalGeneration": (
         "qwen3_vl_moe",