camel-ai 0.2.18__py3-none-any.whl → 0.2.19__py3-none-any.whl

camel/{synthetic_datagen → datagen}/source2synth/models.py

@@ -17,12 +17,30 @@ from pydantic import BaseModel, Field
 
 
 class ReasoningStep(BaseModel):
+    r"""A single step in a multi-hop reasoning process.
+
+    Attributes:
+        step (str): The textual description of the reasoning step.
+    """
+
     step: str = Field(
         ..., description="A single step in the reasoning process."
     )
 
 
 class MultiHopQA(BaseModel):
+    r"""A multi-hop question-answer pair with reasoning steps and supporting
+    facts.
+
+    Attributes:
+        question (str): The question requiring multi-hop reasoning.
+        reasoning_steps (List[ReasoningStep]): List of reasoning steps to
+            answer.
+        answer (str): The final answer to the question.
+        supporting_facts (List[str]): List of facts supporting the reasoning.
+        type (str): The type of question-answer pair.
+    """
+
     question: str = Field(
         ..., description="The question that requires multi-hop reasoning."
     )
@@ -57,6 +75,13 @@ class MultiHopQA(BaseModel):
 
 
 class ContextPrompt(BaseModel):
+    r"""A context prompt for generating multi-hop question-answer pairs.
+
+    Attributes:
+        main_context (str): The primary context for generating QA pairs.
+        related_contexts (Optional[List[str]]): Additional related contexts.
+    """
+
     main_context: str = Field(
         ...,
         description="The main context for generating"

camel/{synthetic_datagen → datagen}/source2synth/user_data_processor_config.py

@@ -23,7 +23,15 @@ class ProcessorConfig(BaseModel):
     r"""Data processing configuration class"""
 
     def __repr__(self):
-        return "MultiHopGeneratorAgent()"
+        return (
+            f"ProcessorConfig("
+            f"seed={self.seed}, min_length={self.min_length}, "
+            f"max_length={self.max_length}, "
+            f"complexity_threshold={self.complexity_threshold}, "
+            f"dataset_size={self.dataset_size}, "
+            f"use_ai_model={self.use_ai_model}"
+            f")"
+        )
 
     model_config = ConfigDict(
         validate_assignment=True,
@@ -45,13 +53,6 @@ class ProcessorConfig(BaseModel):
         default=512, description="Maximum text length", gt=0
     )
 
-    quality_threshold: float = Field(
-        default=0.7,
-        description="Quality threshold for processing",
-        ge=0.0,
-        le=1.0,
-    )
-
     complexity_threshold: float = Field(
         default=0.5,
         description="Complexity threshold for processing",

camel/embeddings/__init__.py

@@ -12,6 +12,7 @@
 # limitations under the License.
 # ========= Copyright 2023-2024 @ CAMEL-AI.org. All Rights Reserved. =========
 from .base import BaseEmbedding
+from .jina_embedding import JinaEmbedding
 from .mistral_embedding import MistralEmbedding
 from .openai_compatible_embedding import OpenAICompatibleEmbedding
 from .openai_embedding import OpenAIEmbedding
@@ -25,4 +26,5 @@ __all__ = [
     "VisionLanguageEmbedding",
     "MistralEmbedding",
     "OpenAICompatibleEmbedding",
+    "JinaEmbedding",
 ]

camel/embeddings/jina_embedding.py

@@ -0,0 +1,156 @@
+# ========= Copyright 2023-2024 @ 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 2023-2024 @ CAMEL-AI.org. All Rights Reserved. =========
+
+import base64
+import io
+import os
+from typing import Any, Optional, Union
+
+import requests
+from PIL import Image
+
+from camel.embeddings import BaseEmbedding
+from camel.types.enums import EmbeddingModelType
+from camel.utils import api_keys_required
+
+
+class JinaEmbedding(BaseEmbedding[Union[str, Image.Image]]):
+    r"""Provides text and image embedding functionalities using Jina AI's API.
+
+    Args:
+        model_type (EmbeddingModelType, optional): The model to use for
+            embeddings. (default: :obj:`JINA_EMBEDDINGS_V3`)
+        api_key (Optional[str], optional): The API key for authenticating with
+            Jina AI. (default: :obj:`None`)
+        dimensions (Optional[int], optional): The dimension of the output
+            embeddings. (default: :obj:`None`)
+        task (Optional[str], optional): The type of task for text embeddings.
+            Options: retrieval.query, retrieval.passage, text-matching,
+            classification, separation. (default: :obj:`None`)
+        late_chunking (bool, optional): If true, concatenates all sentences in
+            input and treats as a single input. (default: :obj:`False`)
+        normalized (bool, optional): If true, embeddings are normalized to unit
+            L2 norm. (default: :obj:`False`)
+    """
+
+    @api_keys_required([("api_key", 'JINA_API_KEY')])
+    def __init__(
+        self,
+        model_type: EmbeddingModelType = EmbeddingModelType.JINA_EMBEDDINGS_V3,
+        api_key: Optional[str] = None,
+        dimensions: Optional[int] = None,
+        embedding_type: Optional[str] = None,
+        task: Optional[str] = None,
+        late_chunking: bool = False,
+        normalized: bool = False,
+    ) -> None:
+        if not model_type.is_jina:
+            raise ValueError(
+                f"Model type {model_type} is not a Jina model. "
+                "Please use a valid Jina model type."
+            )
+        self.model_type = model_type
+        if dimensions is None:
+            self.output_dim = model_type.output_dim
+        else:
+            self.output_dim = dimensions
+        self._api_key = api_key or os.environ.get("JINA_API_KEY")
+
+        self.embedding_type = embedding_type
+        self.task = task
+        self.late_chunking = late_chunking
+        self.normalized = normalized
+        self.url = 'https://api.jina.ai/v1/embeddings'
+        self.headers = {
+            'Content-Type': 'application/json',
+            'Accept': 'application/json',
+            'Authorization': f'Bearer {self._api_key}',
+        }
+
+    def embed_list(
+        self,
+        objs: list[Union[str, Image.Image]],
+        **kwargs: Any,
+    ) -> list[list[float]]:
+        r"""Generates embeddings for the given texts or images.
+
+        Args:
+            objs (list[Union[str, Image.Image]]): The texts or images for which
+                to generate the embeddings.
+            **kwargs (Any): Extra kwargs passed to the embedding API. Not used
+                in this implementation.
+
+        Returns:
+            list[list[float]]: A list that represents the generated embedding
+                as a list of floating-point numbers.
+
+        Raises:
+            ValueError: If the input type is not supported.
+            RuntimeError: If the API request fails.
+        """
+        input_data = []
+        for obj in objs:
+            if isinstance(obj, str):
+                if self.model_type == EmbeddingModelType.JINA_CLIP_V2:
+                    input_data.append({"text": obj})
+                else:
+                    input_data.append(obj)  # type: ignore[arg-type]
+            elif isinstance(obj, Image.Image):
+                if self.model_type != EmbeddingModelType.JINA_CLIP_V2:
+                    raise ValueError(
+                        f"Model {self.model_type} does not support "
+                        "image input. Use JINA_CLIP_V2 for image embeddings."
+                    )
+                # Convert PIL Image to base64 string
+                buffered = io.BytesIO()
+                obj.save(buffered, format="PNG")
+                img_str = base64.b64encode(buffered.getvalue()).decode()
+                input_data.append({"image": img_str})
+            else:
+                raise ValueError(
+                    f"Input type {type(obj)} is not supported. "
+                    "Must be either str or PIL.Image"
+                )
+
+        data = {
+            "model": self.model_type.value,
+            "input": input_data,
+            "embedding_type": "float",
+        }
+
+        if self.embedding_type is not None:
+            data["embedding_type"] = self.embedding_type
+        if self.task is not None:
+            data["task"] = self.task
+        if self.late_chunking:
+            data["late_chunking"] = self.late_chunking  # type: ignore[assignment]
+        if self.normalized:
+            data["normalized"] = self.normalized  # type: ignore[assignment]
+        try:
+            response = requests.post(
+                self.url, headers=self.headers, json=data, timeout=180
+            )
+            response.raise_for_status()
+            result = response.json()
+            return [data["embedding"] for data in result["data"]]
+        except requests.exceptions.RequestException as e:
+            raise RuntimeError(f"Failed to get embeddings from Jina AI: {e}")
+
+    def get_output_dim(self) -> int:
+        r"""Returns the output dimension of the embeddings.
+
+        Returns:
+            int: The dimensionality of the embedding for the current model.
+        """
+        return self.output_dim

camel/messages/func_message.py

@@ -154,7 +154,7 @@ class FunctionCallingMessage(BaseMessage):
                 " due to missing function name."
             )
 
-        result_content = json.dumps(self.result)
+        result_content = str(self.result)
 
         return {
             "role": "tool",

camel/models/deepseek_model.py

@@ -13,12 +13,12 @@
 # ========= Copyright 2023-2024 @ CAMEL-AI.org. All Rights Reserved. =========
 
 import os
-import warnings
 from typing import Any, Dict, List, Optional, Union
 
 from openai import OpenAI, Stream
 
 from camel.configs import DEEPSEEK_API_PARAMS, DeepSeekConfig
+from camel.logger import get_logger
 from camel.messages import OpenAIMessage
 from camel.models.base_model import BaseModelBackend
 from camel.types import (
@@ -28,6 +28,8 @@ from camel.types import (
 )
 from camel.utils import BaseTokenCounter, OpenAITokenCounter, api_keys_required
 
+logger = get_logger(__name__)
+
 
 class DeepSeekModel(BaseModelBackend):
     r"""DeepSeek API in a unified BaseModelBackend interface.
@@ -116,11 +118,12 @@ class DeepSeekModel(BaseModelBackend):
         if self.model_type in [
             ModelType.DEEPSEEK_REASONER,
         ]:
-            warnings.warn(
-                "Warning: You are using an DeepSeek Reasoner model, "
+            import re
+
+            logger.warning(
+                "You are using a DeepSeek Reasoner model, "
                 "which has certain limitations, reference: "
-                "`https://api-docs.deepseek.com/guides/reasoning_model#api-parameters`.",
-                UserWarning,
+                "`https://api-docs.deepseek.com/guides/reasoning_model#api-parameters`"
             )
 
             # Check and remove unsupported parameters and reset the fixed
@@ -138,14 +141,29 @@ class DeepSeekModel(BaseModelBackend):
                 if key in self.model_config_dict:
                     del self.model_config_dict[key]
 
+            # Remove thinking content from messages before sending to API
+            # This ensures only the final response is sent, excluding
+            # intermediate thought processes
+            messages = [
+                {  # type: ignore[misc]
+                    **msg,
+                    'content': re.sub(
+                        r'<think>.*?</think>',
+                        '',
+                        msg['content'],  # type: ignore[arg-type]
+                        flags=re.DOTALL,
+                    ).strip(),
+                }
+                for msg in messages
+            ]
+
         response = self._client.chat.completions.create(
             messages=messages,
             model=self.model_type,
             **self.model_config_dict,
         )
 
-        # Temporary solution to handle the case where
-        # deepseek returns a reasoning_content
+        # Handle reasoning content with <think> tags at the beginning
         if (
             self.model_type
             in [
@@ -156,10 +174,10 @@ class DeepSeekModel(BaseModelBackend):
         ):
             reasoning_content = response.choices[0].message.reasoning_content
             combined_content = (
-                response.choices[0].message.content
-                + "\n\nBELOW IS THE REASONING CONTENT:\n\n"
-                + (reasoning_content if reasoning_content else "")
-            )
+                f"<think>\n{reasoning_content}\n</think>\n"
+                if reasoning_content
+                else ""
+            ) + response.choices[0].message.content
 
             response = ChatCompletion.construct(
                 id=response.id,

camel/models/groq_model.py

@@ -88,8 +88,6 @@ class GroqModel(BaseModelBackend):
             BaseTokenCounter: The token counter following the model's
                 tokenization style.
         """
-        # Make sure you have the access to these open-source model in
-        # HuggingFace
         if not self._token_counter:
             self._token_counter = OpenAITokenCounter(ModelType.GPT_4O_MINI)
         return self._token_counter

camel/models/openai_model.py

@@ -21,7 +21,6 @@ from camel.configs import OPENAI_API_PARAMS, ChatGPTConfig
 from camel.messages import OpenAIMessage
 from camel.models import BaseModelBackend
 from camel.types import (
-    NOT_GIVEN,
     ChatCompletion,
     ChatCompletionChunk,
     ModelType,
@@ -112,6 +111,7 @@ class OpenAIModel(BaseModelBackend):
             ModelType.O1,
             ModelType.O1_MINI,
             ModelType.O1_PREVIEW,
+            ModelType.O3_MINI,
         ]:
             warnings.warn(
                 "Warning: You are using an O1 model (O1_MINI or O1_PREVIEW), "
@@ -148,14 +148,6 @@ class OpenAIModel(BaseModelBackend):
 
             return self._to_chat_completion(response)
 
-        # Removing 'strict': True from the dictionary for
-        # client.chat.completions.create
-        if self.model_config_dict.get('tools') is not NOT_GIVEN:
-            for tool in self.model_config_dict.get('tools', []):
-                function_dict = tool.get('function', {})
-                if 'strict' in function_dict:
-                    del function_dict['strict']
-
         response = self._client.chat.completions.create(
             messages=messages,
             model=self.model_type,

camel/toolkits/search_toolkit.py

@@ -13,10 +13,9 @@
 # ========= Copyright 2023-2024 @ CAMEL-AI.org. All Rights Reserved. =========
 import os
 import xml.etree.ElementTree as ET
-from typing import Any, Dict, List, Literal, Optional, Type, TypeAlias, Union
+from typing import Any, Dict, List, Literal, Optional, TypeAlias, Union
 
 import requests
-from pydantic import BaseModel
 
 from camel.toolkits.base import BaseToolkit
 from camel.toolkits.function_tool import FunctionTool
@@ -77,7 +76,7 @@ class SearchToolkit(BaseToolkit):
         output_type: Literal[
             "searchResults", "sourcedAnswer", "structured"
         ] = "searchResults",
-        structured_output_schema: Union[Type[BaseModel], str, None] = None,
+        structured_output_schema: Optional[str] = None,
     ) -> Dict[str, Any]:
         r"""Search for a query in the Linkup API and return results in various
         formats.
@@ -92,9 +91,9 @@ class SearchToolkit(BaseToolkit):
                 - "searchResults" for raw search results,
                 - "sourcedAnswer" for an answer with supporting sources,
                 - "structured" for output based on a provided schema.
-            structured_output_schema (Union[Type[BaseModel], str, None]): If
-                `output_type` is "structured",specify the schema of the
-                output. Can be a Pydantic BaseModel or a JSON schema string.
+            structured_output_schema (Optional[str]): If `output_type` is
+                "structured", specify the schema of the output. Must be a
+                string representing a valid object JSON schema.
 
         Returns:
             Dict[str, Any]: A dictionary representing the search result. The

camel/types/enums.py

@@ -37,6 +37,7 @@ class ModelType(UnifiedModelType, Enum):
     O1 = "o1"
     O1_PREVIEW = "o1-preview"
     O1_MINI = "o1-mini"
+    O3_MINI = "o3-mini"
 
     GLM_4 = "glm-4"
     GLM_4V = 'glm-4v'
@@ -44,14 +45,11 @@ class ModelType(UnifiedModelType, Enum):
 
     # Groq platform models
     GROQ_LLAMA_3_1_8B = "llama-3.1-8b-instant"
-    GROQ_LLAMA_3_1_70B = "llama-3.1-70b-versatile"
-    GROQ_LLAMA_3_1_405B = "llama-3.1-405b-reasoning"
     GROQ_LLAMA_3_3_70B = "llama-3.3-70b-versatile"
     GROQ_LLAMA_3_3_70B_PREVIEW = "llama-3.3-70b-specdec"
     GROQ_LLAMA_3_8B = "llama3-8b-8192"
     GROQ_LLAMA_3_70B = "llama3-70b-8192"
     GROQ_MIXTRAL_8_7B = "mixtral-8x7b-32768"
-    GROQ_GEMMA_7B_IT = "gemma-7b-it"
     GROQ_GEMMA_2_9B_IT = "gemma2-9b-it"
 
     # TogetherAI platform models support tool calling
@@ -67,6 +65,17 @@ class ModelType(UnifiedModelType, Enum):
     SAMBA_LLAMA_3_1_70B = "Meta-Llama-3.1-70B-Instruct"
     SAMBA_LLAMA_3_1_405B = "Meta-Llama-3.1-405B-Instruct"
 
+    # SGLang models support tool calling
+    SGLANG_LLAMA_3_1_8B = "meta-llama/Meta-Llama-3.1-8B-Instruct"
+    SGLANG_LLAMA_3_1_70B = "meta-llama/Meta-Llama-3.1-70B-Instruct"
+    SGLANG_LLAMA_3_1_405B = "meta-llama/Meta-Llama-3.1-405B-Instruct"
+    SGLANG_LLAMA_3_2_1B = "meta-llama/Llama-3.2-1B-Instruct"
+    SGLANG_MIXTRAL_NEMO = "mistralai/Mistral-Nemo-Instruct-2407"
+    SGLANG_MISTRAL_7B = "mistralai/Mistral-7B-Instruct-v0.3"
+    SGLANG_QWEN_2_5_7B = "Qwen/Qwen2.5-7B-Instruct"
+    SGLANG_QWEN_2_5_32B = "Qwen/Qwen2.5-32B-Instruct"
+    SGLANG_QWEN_2_5_72B = "Qwen/Qwen2.5-72B-Instruct"
+
     STUB = "stub"
 
     # Legacy anthropic models
@@ -190,6 +199,8 @@ class ModelType(UnifiedModelType, Enum):
                 self.is_internlm,
                 self.is_together,
                 self.is_sambanova,
+                self.is_groq,
+                self.is_sglang,
             ]
         )
 
@@ -205,6 +216,7 @@ class ModelType(UnifiedModelType, Enum):
             ModelType.O1,
             ModelType.O1_PREVIEW,
             ModelType.O1_MINI,
+            ModelType.O3_MINI,
         }
 
     @property
@@ -252,14 +264,11 @@ class ModelType(UnifiedModelType, Enum):
         r"""Returns whether this type of models is served by Groq."""
         return self in {
             ModelType.GROQ_LLAMA_3_1_8B,
-            ModelType.GROQ_LLAMA_3_1_70B,
-            ModelType.GROQ_LLAMA_3_1_405B,
             ModelType.GROQ_LLAMA_3_3_70B,
             ModelType.GROQ_LLAMA_3_3_70B_PREVIEW,
             ModelType.GROQ_LLAMA_3_8B,
             ModelType.GROQ_LLAMA_3_70B,
             ModelType.GROQ_MIXTRAL_8_7B,
-            ModelType.GROQ_GEMMA_7B_IT,
             ModelType.GROQ_GEMMA_2_9B_IT,
         }
 
@@ -413,6 +422,20 @@ class ModelType(UnifiedModelType, Enum):
             ModelType.INTERNLM2_PRO_CHAT,
         }
 
+    @property
+    def is_sglang(self) -> bool:
+        return self in {
+            ModelType.SGLANG_LLAMA_3_1_8B,
+            ModelType.SGLANG_LLAMA_3_1_70B,
+            ModelType.SGLANG_LLAMA_3_1_405B,
+            ModelType.SGLANG_LLAMA_3_2_1B,
+            ModelType.SGLANG_MIXTRAL_NEMO,
+            ModelType.SGLANG_MISTRAL_7B,
+            ModelType.SGLANG_QWEN_2_5_7B,
+            ModelType.SGLANG_QWEN_2_5_32B,
+            ModelType.SGLANG_QWEN_2_5_72B,
+        }
+
     @property
     def token_limit(self) -> int:
         r"""Returns the maximum token limit for a given model.
@@ -440,7 +463,6 @@ class ModelType(UnifiedModelType, Enum):
             ModelType.GROQ_LLAMA_3_8B,
             ModelType.GROQ_LLAMA_3_70B,
             ModelType.GROQ_LLAMA_3_3_70B_PREVIEW,
-            ModelType.GROQ_GEMMA_7B_IT,
             ModelType.GROQ_GEMMA_2_9B_IT,
             ModelType.GLM_3_TURBO,
             ModelType.GLM_4,
@@ -479,6 +501,7 @@ class ModelType(UnifiedModelType, Enum):
             ModelType.INTERNLM2_5_LATEST,
             ModelType.INTERNLM2_PRO_CHAT,
             ModelType.TOGETHER_MIXTRAL_8_7B,
+            ModelType.SGLANG_MISTRAL_7B,
         }:
             return 32_768
         elif self in {
@@ -518,12 +541,15 @@ class ModelType(UnifiedModelType, Enum):
             ModelType.NVIDIA_LLAMA3_3_70B_INSTRUCT,
             ModelType.GROQ_LLAMA_3_3_70B,
             ModelType.SAMBA_LLAMA_3_1_70B,
+            ModelType.SGLANG_LLAMA_3_1_8B,
+            ModelType.SGLANG_LLAMA_3_1_70B,
+            ModelType.SGLANG_LLAMA_3_1_405B,
+            ModelType.SGLANG_LLAMA_3_2_1B,
+            ModelType.SGLANG_MIXTRAL_NEMO,
         }:
             return 128_000
         elif self in {
             ModelType.GROQ_LLAMA_3_1_8B,
-            ModelType.GROQ_LLAMA_3_1_70B,
-            ModelType.GROQ_LLAMA_3_1_405B,
             ModelType.QWEN_PLUS,
             ModelType.QWEN_TURBO,
             ModelType.QWEN_CODER_TURBO,
@@ -531,10 +557,14 @@ class ModelType(UnifiedModelType, Enum):
             ModelType.TOGETHER_LLAMA_3_1_70B,
             ModelType.TOGETHER_LLAMA_3_1_405B,
             ModelType.TOGETHER_LLAMA_3_3_70B,
+            ModelType.SGLANG_QWEN_2_5_7B,
+            ModelType.SGLANG_QWEN_2_5_32B,
+            ModelType.SGLANG_QWEN_2_5_72B,
         }:
             return 131_072
         elif self in {
             ModelType.O1,
+            ModelType.O3_MINI,
             ModelType.CLAUDE_2_1,
             ModelType.CLAUDE_3_OPUS,
             ModelType.CLAUDE_3_SONNET,
@@ -567,6 +597,11 @@ class EmbeddingModelType(Enum):
     TEXT_EMBEDDING_3_SMALL = "text-embedding-3-small"
     TEXT_EMBEDDING_3_LARGE = "text-embedding-3-large"
 
+    JINA_EMBEDDINGS_V3 = "jina-embeddings-v3"
+    JINA_CLIP_V2 = "jina-clip-v2"
+    JINA_COLBERT_V2 = "jina-colbert-v2"
+    JINA_EMBEDDINGS_V2_BASE_CODE = "jina-embeddings-v2-base-code"
+
     MISTRAL_EMBED = "mistral-embed"
 
     @property
@@ -578,6 +613,16 @@ class EmbeddingModelType(Enum):
             EmbeddingModelType.TEXT_EMBEDDING_3_LARGE,
         }
 
+    @property
+    def is_jina(self) -> bool:
+        r"""Returns whether this type of models is an Jina model."""
+        return self in {
+            EmbeddingModelType.JINA_EMBEDDINGS_V3,
+            EmbeddingModelType.JINA_CLIP_V2,
+            EmbeddingModelType.JINA_COLBERT_V2,
+            EmbeddingModelType.JINA_EMBEDDINGS_V2_BASE_CODE,
+        }
+
     @property
     def is_mistral(self) -> bool:
         r"""Returns whether this type of models is an Mistral-released
@@ -589,7 +634,20 @@ class EmbeddingModelType(Enum):
 
     @property
     def output_dim(self) -> int:
-        if self is EmbeddingModelType.TEXT_EMBEDDING_ADA_2:
+        if self in {
+            EmbeddingModelType.JINA_COLBERT_V2,
+        }:
+            return 128
+        elif self in {
+            EmbeddingModelType.JINA_EMBEDDINGS_V2_BASE_CODE,
+        }:
+            return 768
+        elif self in {
+            EmbeddingModelType.JINA_EMBEDDINGS_V3,
+            EmbeddingModelType.JINA_CLIP_V2,
+        }:
+            return 1024
+        elif self is EmbeddingModelType.TEXT_EMBEDDING_ADA_2:
             return 1536
         elif self is EmbeddingModelType.TEXT_EMBEDDING_3_SMALL:
             return 1536

camel/utils/token_counting.py

@@ -112,7 +112,7 @@ class OpenAITokenCounter(BaseTokenCounter):
         elif ("gpt-3.5-turbo" in self.model) or ("gpt-4" in self.model):
             self.tokens_per_message = 3
             self.tokens_per_name = 1
-        elif "o1" in self.model:
+        elif ("o1" in self.model) or ("o3" in self.model):
             self.tokens_per_message = 2
             self.tokens_per_name = 1
         else:

{camel_ai-0.2.18.dist-info → camel_ai-0.2.19.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: camel-ai
-Version: 0.2.18
+Version: 0.2.19
 Summary: Communicative Agents for AI Society Study
 License: Apache-2.0
 Keywords: communicative-ai,ai-societies,artificial-intelligence,deep-learning,multi-agent-systems,cooperative-ai,natural-language-processing,large-language-models
@@ -71,7 +71,7 @@ Requires-Dist: neo4j (>=5.18.0,<6.0.0) ; extra == "rag" or extra == "storage" or
 Requires-Dist: newspaper3k (>=0.2.8,<0.3.0) ; extra == "web-tools" or extra == "all"
 Requires-Dist: notion-client (>=2.2.1,<3.0.0) ; extra == "communication-tools" or extra == "all"
 Requires-Dist: numpy (>=1,<2)
-Requires-Dist: openai (>=1.58.1,<2.0.0)
+Requires-Dist: openai (>=1.59.7,<2.0.0)
 Requires-Dist: openapi-spec-validator (>=0.7.1,<0.8.0) ; extra == "document-tools" or extra == "all"
 Requires-Dist: openbb (>=4.3.5,<5.0.0) ; extra == "data-tools" or extra == "all"
 Requires-Dist: opencv-python (>=4,<5) ; extra == "huggingface" or extra == "all"
@@ -418,6 +418,7 @@ Practical guides and tutorials for implementing specific functionalities in CAME
 | **[Dynamic Travel Planner Role-Playing: Multi-Agent System with Real-Time Insights Powered by Dappier](https://docs.camel-ai.org/cookbooks/applications/dynamic_travel_planner.html)** | Explore an innovative approach to travel planning, blending AI-driven role-playing and real-time data for seamless experiences. |
 | **[Customer Service Discord Bot with Agentic RAG](https://docs.camel-ai.org/cookbooks/applications/customer_service_Discord_bot_using_SambaNova_with_agentic_RAG.html)** | Learn how to build a robust customer service bot for Discord using Agentic RAG. |
 | **[Customer Service Discord Bot with Local Model](https://docs.camel-ai.org/cookbooks/applications/customer_service_Discord_bot_using_local_model_with_agentic_RAG.html)** | Learn how to build a robust customer service bot for Discord using Agentic RAG which supports local deployment. |
+| **[Customer Service Discord Bot for Finance with OpenBB](https://docs.camel-ai.org/cookbooks/applications/finance_discord_bot.html)**| Learn how to build a sipmle yet powerful financial data assistant Discord bot using OpenBB tools. |
 
 ### Data Processing
 | Cookbook | Description |
@@ -456,6 +457,8 @@ We implemented amazing research ideas from other works for you to build, compare
 
 - `Self-Instruct` from *Yizhong Wang et al.*: [SELF-INSTRUCT: Aligning Language Models with Self-Generated Instructions](https://arxiv.org/pdf/2212.10560). [[Example](https://github.com/camel-ai/camel/blob/master/examples/datagen/self_instruct/self_instruct.py)]
 
+- `Source2Synth` from *Alisia Lupidi et al.*: [Source2Synth: Synthetic Data Generation and Curation Grounded in Real Data Sources](https://arxiv.org/abs/2409.08239). [[Example](https://github.com/camel-ai/camel/blob/master/examples/datagen/source2synth.py)]
+
 ## Other Research Works Based on Camel
 - [Agent Trust](http://agent-trust.camel-ai.org/): Can Large Language Model Agents Simulate Human Trust Behavior?