#                🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨
#           This file was automatically generated from src/transformers/models/minicpmv4_6/modular_minicpmv4_6.py.
#               Do NOT edit this file manually as any edits will be overwritten by the generation of
#             the file from the modular. If any change should be done, please apply the change to the
#                          modular_minicpmv4_6.py file directly. One of our CI enforces this.
#                🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨
# Copyright 2026 OpenBMB and the HuggingFace Inc. team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.


from huggingface_hub.dataclasses import strict

from ...configuration_utils import PreTrainedConfig
from ...utils import auto_docstring
from ..auto import CONFIG_MAPPING, AutoConfig


@auto_docstring(checkpoint="openbmb/MiniCPM-V-4.6")
@strict
class MiniCPMV4_6VisionConfig(PreTrainedConfig):
    r"""
    insert_layer_id (`int`, *optional*, defaults to 6):
        Vision encoder layer index after which the window-attention merger is applied.
    window_kernel_size (`tuple[int, int]`, *optional*, defaults to `(2, 2)`):
        Window size `(h, w)` for the intermediate window-attention merger.
    """

    model_type = "minicpmv4_6_vision"
    base_config_key = "vision_config"

    hidden_size: int = 768
    intermediate_size: int = 3072
    num_hidden_layers: int = 12
    num_attention_heads: int = 12
    num_channels: int = 3
    image_size: int | list[int] | tuple[int, int] = 224
    patch_size: int | list[int] | tuple[int, int] = 16
    hidden_act: str = "gelu_pytorch_tanh"
    layer_norm_eps: float = 1e-6
    attention_dropout: float | int = 0.0
    insert_layer_id: int = 6
    window_kernel_size: tuple[int, int] | list[int] = (2, 2)

    @property
    def window_hidden_size(self) -> int:
        return self.hidden_size * self.window_kernel_size[0] * self.window_kernel_size[1]

    @property
    def window_intermediate_size(self) -> int:
        return self.intermediate_size * self.window_kernel_size[0] * self.window_kernel_size[1]


@auto_docstring(checkpoint="openbmb/MiniCPM-V-4.6")
@strict
class MiniCPMV4_6Config(PreTrainedConfig):
    r"""
    insert_layer_id (`int`, *optional*, defaults to 6):
        Vision encoder layer index after which the window-attention merger is applied.
    image_size (`int`, *optional*, defaults to 448):
        Base resolution for image preprocessing.
    drop_vision_last_layer (`bool`, *optional*, defaults to `False`):
        Whether to drop the last layer of the vision encoder.
    image_token_id (`int`, *optional*):
        Token id used as the image placeholder.
    video_token_id (`int`, *optional*):
        Token id used as the video placeholder.
    downsample_mode (`str`, *optional*, defaults to `"16x"`):
        Visual token downsampling ratio. `"4x"` keeps 4× more tokens.
    merge_kernel_size (`tuple[int, int]`, *optional*, defaults to `(2, 2)`):
        Kernel size `(h, w)` for merging adjacent visual patches in the Merger.
    merger_times (`int`, *optional*, defaults to 1):
        Number of iterative merge rounds in the Merger.
    """

    model_type = "minicpmv4_6"
    sub_configs = {"text_config": AutoConfig, "vision_config": MiniCPMV4_6VisionConfig}

    text_config: dict | PreTrainedConfig | None = None
    vision_config: dict | PreTrainedConfig | None = None
    insert_layer_id: int = 6
    image_size: int = 448
    drop_vision_last_layer: bool = False
    image_token_id: int | None = None
    video_token_id: int | None = None
    tie_word_embeddings: bool = False
    downsample_mode: str = "16x"
    merge_kernel_size: tuple[int, int] | list[int] = (2, 2)
    merger_times: int = 1

    def __post_init__(self, **kwargs):
        if isinstance(self.vision_config, dict):
            self.vision_config.pop("model_type", None)
            self.vision_config = MiniCPMV4_6VisionConfig(**self.vision_config)
        elif self.vision_config is None:
            self.vision_config = MiniCPMV4_6VisionConfig()

        self.vision_config.insert_layer_id = self.insert_layer_id
        self.patch_size = self.vision_config.patch_size

        if isinstance(self.text_config, dict):
            self.text_config = CONFIG_MAPPING[self.text_config["model_type"]](**self.text_config)
        elif self.text_config is None:
            self.text_config = CONFIG_MAPPING["qwen3_5_text"]()

        super().__post_init__(**kwargs)


__all__ = ["MiniCPMV4_6Config", "MiniCPMV4_6VisionConfig"]