#                🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨
#           This file was automatically generated from src/transformers/models/exaone4_5/modular_exaone4_5.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_exaone4_5.py file directly. One of our CI enforces this.
#                🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨
# Copyright 2026 The LG AI Research and 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="LGAI-EXAONE/EXAONE-4.5-33B")
@strict
class Exaone4_5_VisionConfig(PreTrainedConfig):
    r"""
    tokens_per_second (`int`, *optional*, defaults to 41):
        Number of tokens to merge for each second of video.
    window_size (`int`, *optional*, defaults to 11):
        Size of windows.
    out_hidden_size (`int`, *optional*, defaults to 3584):
        The output hidden size of the vision model.
    fullatt_block_indexes (`int`, *optional*, defaults to `[7, 15, 23, 31]`):
        Indices of layers with full attention
    """

    model_type = "exaone4_5_vision"
    base_config_key = "vision_config"

    depth: int = 32
    hidden_size: int = 3584
    hidden_act: str = "silu"
    intermediate_size: int = 3420
    num_heads: int = 16
    in_channels: int = 3
    patch_size: int | list[int] | tuple[int, int] = 14
    spatial_merge_size: int = 2
    temporal_patch_size: int | list[int] | tuple[int, int] = 2
    tokens_per_second: int = 4
    window_size: int = 112
    out_hidden_size: int = 3584
    fullatt_block_indexes: list[int] | tuple[int, ...] = (7, 15, 23, 31)
    initializer_range: float = 0.02
    num_key_value_heads: int = 8


@auto_docstring(checkpoint="LGAI-EXAONE/EXAONE-4.5-33B")
@strict
class Exaone4_5_Config(PreTrainedConfig):
    model_type = "exaone4_5"
    sub_configs = {"vision_config": AutoConfig, "text_config": AutoConfig}
    keys_to_ignore_at_inference = ["past_key_values"]

    text_config: dict | PreTrainedConfig | None = None
    vision_config: dict | PreTrainedConfig | None = None
    image_token_id: int = 67
    video_token_id: int = 68
    tie_word_embeddings: bool = False

    def __post_init__(self, **kwargs):
        if isinstance(self.vision_config, dict):
            self.vision_config["model_type"] = self.vision_config.get("model_type", "exaone4_5_vision")
            self.vision_config = CONFIG_MAPPING[self.vision_config["model_type"]](**self.vision_config)
        elif self.vision_config is None:
            self.vision_config = CONFIG_MAPPING["exaone4_5_vision"]()

        if isinstance(self.text_config, dict):
            self.text_config["model_type"] = self.text_config.get("model_type", "exaone4")
            # BC: EXAONE 4.5 first released with the text model type as `exaone4_5_text`, now changed to `exaone4`
            if self.text_config["model_type"] == "exaone4_5_text":
                self.text_config["model_type"] = "exaone4"
            self.text_config = CONFIG_MAPPING[self.text_config["model_type"]](**self.text_config)
        elif self.text_config is None:
            self.text_config = CONFIG_MAPPING["exaone4"]()

        super().__post_init__(**kwargs)


__all__ = ["Exaone4_5_Config", "Exaone4_5_VisionConfig"]