Skip to content

vllm.model_executor.models.opencua

Inference-only OpenCUA-7B model compatible with HuggingFace weights.

Classes:

OpenCUAProcessingInfo

Bases: Qwen2VLProcessingInfo

Methods:

Source code in vllm/model_executor/models/opencua.py 
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
class OpenCUAProcessingInfo(Qwen2VLProcessingInfo):
    def get_data_parser(self):
        return Qwen2VLMultiModalDataParser(
            self.get_hf_config().vision_config.spatial_merge_size,
            expected_hidden_size=self._get_expected_hidden_size(),
            allow_missing_mm_embeddings=self.allow_missing_mm_embeddings,
        )

    def get_hf_config(self):
        return self.ctx.get_hf_config()

    def get_supported_mm_limits(self) -> Mapping[str, int | None]:
        return {"image": None}

    def get_hf_processor(self, **kwargs: object):
        """Load OpenCUA processor."""
        tokenizer = self.get_tokenizer()
        vision_config = self.ctx.get_hf_image_processor_config()
        return OpenCUAProcessor(
            vision_config=vision_config,
            tokenizer=tokenizer,
            **kwargs,
        )

get_hf_processor(**kwargs)

Load OpenCUA processor.

Source code in vllm/model_executor/models/opencua.py 
69
70
71
72
73
74
75
76
77
def get_hf_processor(self, **kwargs: object):
    """Load OpenCUA processor."""
    tokenizer = self.get_tokenizer()
    vision_config = self.ctx.get_hf_image_processor_config()
    return OpenCUAProcessor(
        vision_config=vision_config,
        tokenizer=tokenizer,
        **kwargs,
    )