raise NotImplementedError for num_reqs > 1 and run precommit

Signed-off-by: HonestDeng <2958906959@qq.com>

Author: HonestDeng

tests/e2e/offline_inference/test_mammoth_moda2.py

@@ -48,7 +48,9 @@
 
 _EXAMPLE_DIR = Path(__file__).resolve().parents[3] / "examples" / "offline_inference" / "mammothmodal2_preview"
 _STAGE_CONFIGS_DIR = Path(__file__).resolve().parents[3] / "vllm_omni" / "model_executor" / "stage_configs"
-MODEL_PATH = os.environ.get("MAMMOTHMODA2_MODEL_PATH", str(Path(__file__).resolve().parents[3] / "MammothModa2-Preview"))
+MODEL_PATH = os.environ.get(
+    "MAMMOTHMODA2_MODEL_PATH", str(Path(__file__).resolve().parents[3] / "MammothModa2-Preview")
+)
 T2I_STAGE_CONFIG = os.environ.get("MAMMOTHMODA2_T2I_STAGE_CONFIG", str(_STAGE_CONFIGS_DIR / "mammoth_moda2.yaml"))
 SUMMARIZE_STAGE_CONFIG = os.environ.get(
     "MAMMOTHMODA2_SUMMARIZE_STAGE_CONFIG", str(_STAGE_CONFIGS_DIR / "mammoth_moda2_ar.yaml")

vllm_omni/__init__.py

@@ -23,7 +23,6 @@
 # Register custom configs (AutoConfig, AutoTokenizer) as early as possible.
 from vllm_omni.transformers_utils import configs as _configs  # noqa: F401, E402
 
-
 from .config import OmniModelConfig
 from .entrypoints.async_omni import AsyncOmni
 

vllm_omni/diffusion/models/mammoth_moda2/mammoth_moda2_dit.py

@@ -11,8 +11,8 @@
 from vllm.config import VllmConfig
 from vllm.model_executor.models.utils import AutoWeightsLoader, WeightsMapper
 
-from vllm_omni.transformers_utils.configs.mammoth_moda2 import Mammothmoda2Config
 from vllm_omni.model_executor.models.output_templates import OmniOutput
+from vllm_omni.transformers_utils.configs.mammoth_moda2 import Mammothmoda2Config
 
 from .mammothmoda2_dit_model import SimpleQFormerImageRefiner, Transformer2DModel
 from .rope_real import RotaryPosEmbedReal
@@ -93,9 +93,12 @@ def _reinit_caption_embedder(self, in_features: int) -> None:
         )
 
     def get_dummy_runtime_additional_information(self, num_reqs: int) -> list[dict[str, object]]:
-        num_reqs = 1  # TODO: support num_reqs > 1
         if num_reqs <= 0:
             raise ValueError(f"num_reqs must be positive, got {num_reqs}")
+        if num_reqs > 1:
+            raise NotImplementedError(
+                f"get_dummy_runtime_additional_information does not support num_reqs > 1, got {num_reqs}"
+            )
         text_prompt_embeds = torch.zeros((1, self._llm_hidden_size), dtype=torch.float32)
         image_prompt_embeds = torch.zeros((1, self._llm_hidden_size), dtype=torch.float32)
         negative_prompt_embeds = torch.zeros((0, self._llm_hidden_size), dtype=torch.float32)

vllm_omni/model_executor/models/mammoth_moda2/mammoth_moda2.py

@@ -408,7 +408,7 @@ def _make_empty_intermediate_tensors(
             self.norm = PPMissingLayer()
 
     @property
-    def model(self) -> "MammothModa2Qwen2ForCausalLM":
+    def model(self) -> MammothModa2Qwen2ForCausalLM:
         return self
 
     def get_input_embeddings(self, input_ids: torch.Tensor) -> torch.Tensor: