feat: Implement CogVideo model

camel/models/__init__.py

@@ -14,6 +14,7 @@
 from .anthropic_model import AnthropicModel
 from .azure_openai_model import AzureOpenAIModel
 from .base_model import BaseModelBackend
+from .cogvideo_model import CogVideoModel
 from .gemini_model import GeminiModel
 from .groq_model import GroqModel
 from .litellm_model import LiteLLMModel
@@ -53,4 +54,5 @@
     'RekaModel',
     'SambaModel',
     'TogetherAIModel',
+    'CogVideoModel',
 ]

camel/models/cogvideo_model.py

@@ -0,0 +1,90 @@
+# =========== Copyright 2023 @ CAMEL-AI.org. 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.
+# =========== Copyright 2023 @ CAMEL-AI.org. All Rights Reserved. ===========
+
+import os
+from typing import Any, Dict, Optional
+
+import httpx
+import requests
+
+from camel.types import ModelType
+
+
+class CogVideoModel:
+    r"""CogVideo model API backend."""
+
+    def __init__(
+        self,
+        model_type: ModelType,
+        model_config_dict: Dict[str, Any],
+        url: Optional[str] = "http://localhost:8000/generate",
+        use_gpu: bool = True,
+    ) -> None:
+        r"""Constructor for CogVideo backend
+
+        Reference: https://github.com/THUDM/CogVideo
+
+        Args:
+            model_type (ModelType): Model for which backend is created
+                such as CogVideoX-2B, CogVideoX-5B, etc.
+            model_config_dict (Dict[str, Any]): A dictionary of parameters
+                for the model configuration.
+            url (Optional[str]): The URL to the model service.
+                (default: 'http://localhost:8000/generate')
+            use_gpu (bool): Whether to use GPU for inference. (default: True)
+        """
+        self.model_type = model_type
+        self.model_config_dict = model_config_dict
+        self._url = url or os.environ.get("COGVIDEO_API_BASE_URL")
+        if not self._url:
+            raise ValueError("COGVIDEO_API_BASE_URL should be set.")
+        self._use_gpu = use_gpu
+
+    async def run(self, prompt: str, **kwargs: Any) -> str:
+        r"""Run the CogVideo model to generate a video from a text prompt.
+
+        Args:
+            prompt (str): The text prompt to generate the video.
+            **kwargs (Any): Additional arguments for the model request.
+
+        Returns:
+            str: The path or URL to the generated video.
+
+        Raises:
+            Exception: If there is an error in the request or response.
+        """
+        data = {
+            "prompt": prompt,
+            "model_type": self.model_type,
+            "use_gpu": self._use_gpu,
+            **self.model_config_dict,
+            **kwargs,
+        }
+
+        if not isinstance(self._url, str):
+            raise ValueError("URL should be a string.")
+
+        async with httpx.AsyncClient() as client:
+            try:
+                response = await client.post(self._url, json=data)
+                response.raise_for_status()
+                video_url = response.json().get("video_url")
+                if not video_url:
+                    raise ValueError(
+                        "No video URL returned by the model service."
+                    )
+                return video_url
+
+            except requests.exceptions.RequestException as e:
+                raise Exception("Error during CogVideo API call") from e

camel/types/enums.py

@@ -83,6 +83,11 @@ class ModelType(Enum):
     MISTRAL_MIXTRAL_8x22B = "open-mixtral-8x22b"
     MISTRAL_CODESTRAL_MAMBA = "open-codestral-mamba"
 
+    # CogVideo Model
+    COGVIDEO = "cogvideo"
+    COGVIDEOX_2B = "cogvideox-2b"
+    COGVIDEOX_5B = "cogvideox-5b"
+
     # Reka models
     REKA_CORE = "reka-core"
     REKA_FLASH = "reka-flash"
@@ -209,6 +214,17 @@ def is_gemini(self) -> bool:
         return self in {ModelType.GEMINI_1_5_FLASH, ModelType.GEMINI_1_5_PRO}
 
     @property
+    def is_cogvideo(self) -> bool:
+        r"""Returns whether this type of models is CogVideo-released model.
+
+        Returns:
+            bool: Whether this type of models is CogVideo.
+        """
+        return self in {
+            ModelType.COGVIDEO,
+            ModelType.COGVIDEOX_2B,
+            ModelType.COGVIDEOX_5B,
+
     def is_reka(self) -> bool:
         r"""Returns whether this type of models is Reka model.
 
@@ -395,6 +411,7 @@ class TaskType(Enum):
     MULTI_CONDITION_IMAGE_CRAFT = "multi_condition_image_craft"
     DEFAULT = "default"
     VIDEO_DESCRIPTION = "video_description"
+    VIDEO_GENERATION = "video_generation"
 
 
 class VectorDistance(Enum):
@@ -481,6 +498,8 @@ class ModelPlatformType(Enum):
     REKA = "reka"
     TOGETHER = "together"
     OPENAI_COMPATIBILITY_MODEL = "openai-compatibility-model"
+    INTERNLM = "internlm"
+    COGVIDEO = "cogvideo"
     SAMBA = "samba-nova"
 
     @property
@@ -559,6 +578,11 @@ def is_samba(self) -> bool:
         r"""Returns whether this platform is Samba Nova."""
         return self is ModelPlatformType.SAMBA
 
+    @property
+    def is_cogvideo(self) -> bool:
+        r"""Returns whether this platform is CogVideo"""
+        return self in [ModelPlatformType.COGVIDEO]
+
 
 class AudioModelType(Enum):
     TTS_1 = "tts-1"

examples/models/cogvideo_model_example.py

@@ -0,0 +1,44 @@
+# =========== Copyright 2023 @ CAMEL-AI.org. 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.
+# =========== Copyright 2023 @ CAMEL-AI.org. All Rights Reserved. ===========
+
+import asyncio
+
+from camel.models import CogVideoModel
+from camel.types import ModelType
+
+
+async def main():
+    # Initialize the CogVideo model
+    model = CogVideoModel(
+        model_type=ModelType.COGVIDEOX_5B,
+        model_config_dict={"video_length": 4, "frame_rate": 8},
+    )
+
+    # Define the text prompt fro video generation
+    prompt = "A video of a cat playing with a ball."
+
+    # Run the model to generate video from the text prompt
+    try:
+        video_url = await model.run(prompt=prompt)
+        print(f"Generated video URL: {video_url}")
+    except Exception as e:
+        print(f"An error occurred: {e}")
+
+
+if __name__ == "__main__":
+    # Note: To run this code, you need to have a CogVideo server running
+    # locally or remotely that provides endpoints for this model.
+    # Ensure that the server is accessible at the specified URL
+    # in the CogVideoModel class.
+    asyncio.run(main())