Skip to content

vllm.transformers_utils.configs.kimi_vl

KimiVLConfig

Bases: PretrainedConfig

Source code in vllm/transformers_utils/configs/kimi_vl.py 
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
class KimiVLConfig(PretrainedConfig):
    model_type = "kimi_vl"

    def __init__(self,
                 vision_config: Optional[Union[dict, MoonViTConfig]] = None,
                 text_config: Optional[Union[dict, DeepseekV2Config]] = None,
                 ignore_index: int = -100,
                 media_placeholder_token_id: int = 163605,
                 pad_token_id: int = 0,
                 **kwargs):
        if vision_config is None:
            vision_config = MoonViTConfig()
        elif isinstance(vision_config, dict):
            vision_config = MoonViTConfig(**vision_config)
        self.vision_config = vision_config

        if text_config is None:
            text_config = DeepseekV2Config()
        elif isinstance(text_config, dict):
            text_config = DeepseekV2Config(**text_config)
        self.text_config = text_config

        self.ignore_index = ignore_index
        self.media_placeholder_token_id = media_placeholder_token_id

        super().__init__(pad_token_id=pad_token_id, **kwargs)

ignore_index instance-attribute 

ignore_index = ignore_index

media_placeholder_token_id instance-attribute 

media_placeholder_token_id = media_placeholder_token_id

model_type class-attribute instance-attribute 

model_type = 'kimi_vl'

text_config instance-attribute 

text_config = text_config

vision_config instance-attribute 

vision_config = vision_config

__init__ 

__init__(
    vision_config: Optional[
        Union[dict, MoonViTConfig]
    ] = None,
    text_config: Optional[
        Union[dict, DeepseekV2Config]
    ] = None,
    ignore_index: int = -100,
    media_placeholder_token_id: int = 163605,
    pad_token_id: int = 0,
    **kwargs,
)
Source code in vllm/transformers_utils/configs/kimi_vl.py 
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
def __init__(self,
             vision_config: Optional[Union[dict, MoonViTConfig]] = None,
             text_config: Optional[Union[dict, DeepseekV2Config]] = None,
             ignore_index: int = -100,
             media_placeholder_token_id: int = 163605,
             pad_token_id: int = 0,
             **kwargs):
    if vision_config is None:
        vision_config = MoonViTConfig()
    elif isinstance(vision_config, dict):
        vision_config = MoonViTConfig(**vision_config)
    self.vision_config = vision_config

    if text_config is None:
        text_config = DeepseekV2Config()
    elif isinstance(text_config, dict):
        text_config = DeepseekV2Config(**text_config)
    self.text_config = text_config

    self.ignore_index = ignore_index
    self.media_placeholder_token_id = media_placeholder_token_id

    super().__init__(pad_token_id=pad_token_id, **kwargs)