Skip to content

vllm.v1.worker.xpu_model_runner

Classes:

XPUMMEncoderModelRunner

Bases: MMEncoderModelRunner

An encoder-only model runner for XPU devices.

Source code in vllm/v1/worker/xpu_model_runner.py 
43
44
45
46
47
48
49
50
51
52
class XPUMMEncoderModelRunner(MMEncoderModelRunner):
    """An encoder-only model runner for XPU devices."""

    def __init__(
        self,
        vllm_config: VllmConfig,
        device: torch.device,
    ):
        with _torch_cuda_wrapper():
            super().__init__(vllm_config, device)

XPUModelRunner

Bases: GPUModelRunner

A model runner for XPU devices.

Source code in vllm/v1/worker/xpu_model_runner.py 
17
18
19
20
21
22
23
24
25
26
27
28
class XPUModelRunner(GPUModelRunner):
    """A model runner for XPU devices."""

    def __init__(
        self,
        vllm_config: VllmConfig,
        device: torch.device,
    ):
        with _torch_cuda_wrapper():
            super().__init__(vllm_config, device)
        # FIXME: To be verified.
        self.cascade_attn_enabled = False

XPUModelRunnerV2

Bases: GPUModelRunner

A model runner for XPU devices.

Source code in vllm/v1/worker/xpu_model_runner.py 
31
32
33
34
35
36
37
38
39
40
class XPUModelRunnerV2(GPUModelRunnerV2):
    """A model runner for XPU devices."""

    def __init__(
        self,
        vllm_config: VllmConfig,
        device: torch.device,
    ):
        with _torch_cuda_wrapper():
            super().__init__(vllm_config, device)