Skip to content

armgit / camel

Public Imported from GitHub
0

🐫 CAMEL: The first and the best multi-agent framework. Finding the Scaling Law of Agents. https://www.camel-ai.org

feat: Add a new llm provider - novita (#2284)

Neal Davis103a34cf7089
diff --git a/.github/workflows/build_package.yml b/.github/workflows/build_package.yml
index 5e0c57a1b..fc084a1f5 100644
--- a/.github/workflows/build_package.yml
+++ b/.github/workflows/build_package.yml
@@ -107,6 +107,7 @@ jobs:
BEDROCK_API_KEY: "${{ secrets.BEDROCK_API_KEY }}"
BEDROCK_API_BASE_URL: "${{ secrets.BEDROCK_API_BASE_URL }}"
NETMIND_API_KEY: "${{ secrets.NETMIND_API_KEY }}"
+ NOVITA_API_KEY: "${{ secrets.NOVITA_API_KEY }}"
run: |
source test_venv/bin/activate
pytest --fast-test-mode ./test
diff --git a/.github/workflows/pytest_package.yml b/.github/workflows/pytest_package.yml
index f872c4d0d..bd64419c7 100644
--- a/.github/workflows/pytest_package.yml
+++ b/.github/workflows/pytest_package.yml
@@ -72,6 +72,7 @@ jobs:
BEDROCK_API_KEY: "${{ secrets.BEDROCK_API_KEY }}"
BEDROCK_API_BASE_URL: "${{ secrets.BEDROCK_API_BASE_URL }}"
NETMIND_API_KEY: "${{ secrets.NETMIND_API_KEY }}"
+ NOVITA_API_KEY: "${{ secrets.NOVITA_API_KEY }}"
run: |
source .venv/bin/activate
uv pip install -e ".[all, dev, docs]"
@@ -136,6 +137,7 @@ jobs:
BEDROCK_API_KEY: "${{ secrets.BEDROCK_API_KEY }}"
BEDROCK_API_BASE_URL: "${{ secrets.BEDROCK_API_BASE_URL }}"
NETMIND_API_KEY: "${{ secrets.NETMIND_API_KEY }}"
+ NOVITA_API_KEY: "${{ secrets.NOVITA_API_KEY }}"
run: |
source .venv/bin/activate
uv pip install -e ".[all, dev, docs]"
@@ -198,6 +200,7 @@ jobs:
BEDROCK_API_KEY: "${{ secrets.BEDROCK_API_KEY }}"
BEDROCK_API_BASE_URL: "${{ secrets.BEDROCK_API_BASE_URL }}"
NETMIND_API_KEY: "${{ secrets.NETMIND_API_KEY }}"
+ NOVITA_API_KEY: "${{ secrets.NOVITA_API_KEY }}"
run: |
source .venv/bin/activate
uv pip install -e ".[all, dev, docs]"
diff --git a/camel/configs/__init__.py b/camel/configs/__init__.py
index 45fd107a2..55225172b 100644
--- a/camel/configs/__init__.py
+++ b/camel/configs/__init__.py
@@ -26,6 +26,7 @@ from .mistral_config import MISTRAL_API_PARAMS, MistralConfig
from .modelscope_config import MODELSCOPE_API_PARAMS, ModelScopeConfig
from .moonshot_config import MOONSHOT_API_PARAMS, MoonshotConfig
from .netmind_config import NETMIND_API_PARAMS, NetmindConfig
+from .novita_config import NOVITA_API_PARAMS, NovitaConfig
from .nvidia_config import NVIDIA_API_PARAMS, NvidiaConfig
from .ollama_config import OLLAMA_API_PARAMS, OllamaConfig
from .openai_config import OPENAI_API_PARAMS, ChatGPTConfig
@@ -98,6 +99,8 @@ __all__ = [
"MOONSHOT_API_PARAMS",
'ModelScopeConfig',
'MODELSCOPE_API_PARAMS',
+ 'NovitaConfig',
+ 'NOVITA_API_PARAMS',
'SiliconFlowConfig',
'SILICONFLOW_API_PARAMS',
'AIMLConfig',
diff --git a/camel/configs/novita_config.py b/camel/configs/novita_config.py
new file mode 100644
index 000000000..a55d8e2f9
--- /dev/null
+++ b/camel/configs/novita_config.py
@@ -0,0 +1,102 @@
+# ========= 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. =========
+from __future__ import annotations
+
+from typing import Dict, Optional, Sequence, Type, Union
+
+from pydantic import BaseModel
+
+from camel.configs.base_config import BaseConfig
+
+
+class NovitaConfig(BaseConfig):
+ r"""Defines the parameters for generating chat completions using the
+ OpenAI API.
+
+ Args:
+ temperature (float, optional): Sampling temperature to use, between
+ :obj:`0` and :obj:`2`. Higher values make the output more random,
+ while lower values make it more focused and deterministic.
+ (default: :obj:`None`)
+ top_p (float, optional): An alternative to sampling with temperature,
+ called nucleus sampling, where the model considers the results of
+ the tokens with top_p probability mass. So :obj:`0.1` means only
+ the tokens comprising the top 10% probability mass are considered.
+ (default: :obj:`None`)
+ n (int, optional): How many chat completion choices to generate for
+ each input message. (default: :obj:`None`)
+ response_format (object, optional): An object specifying the format
+ that the model must output. Compatible with GPT-4 Turbo and all
+ GPT-3.5 Turbo models newer than gpt-3.5-turbo-1106. Setting to
+ {"type": "json_object"} enables JSON mode, which guarantees the
+ message the model generates is valid JSON. Important: when using
+ JSON mode, you must also instruct the model to produce JSON
+ yourself via a system or user message. Without this, the model
+ may generate an unending stream of whitespace until the generation
+ reaches the token limit, resulting in a long-running and seemingly
+ "stuck" request. Also note that the message content may be
+ partially cut off if finish_reason="length", which indicates the
+ generation exceeded max_tokens or the conversation exceeded the
+ max context length.
+ stream (bool, optional): If True, partial message deltas will be sent
+ as data-only server-sent events as they become available.
+ (default: :obj:`None`)
+ stop (str or list, optional): Up to :obj:`4` sequences where the API
+ will stop generating further tokens. (default: :obj:`None`)
+ max_tokens (int, optional): The maximum number of tokens to generate
+ in the chat completion. The total length of input tokens and
+ generated tokens is limited by the model's context length.
+ (default: :obj:`None`)
+ presence_penalty (float, optional): Number between :obj:`-2.0` and
+ :obj:`2.0`. Positive values penalize new tokens based on whether
+ they appear in the text so far, increasing the model's likelihood
+ to talk about new topics. See more information about frequency and
+ presence penalties. (default: :obj:`None`)
+ frequency_penalty (float, optional): Number between :obj:`-2.0` and
+ :obj:`2.0`. Positive values penalize new tokens based on their
+ existing frequency in the text so far, decreasing the model's
+ likelihood to repeat the same line verbatim. See more information
+ about frequency and presence penalties. (default: :obj:`None`)
+ logit_bias (dict, optional): Modify the likelihood of specified tokens
+ appearing in the completion. Accepts a json object that maps tokens
+ (specified by their token ID in the tokenizer) to an associated
+ bias value from :obj:`-100` to :obj:`100`. Mathematically, the bias
+ is added to the logits generated by the model prior to sampling.
+ The exact effect will vary per model, but values between:obj:` -1`
+ and :obj:`1` should decrease or increase likelihood of selection;
+ values like :obj:`-100` or :obj:`100` should result in a ban or
+ exclusive selection of the relevant token. (default: :obj:`None`)
+ user (str, optional): A unique identifier representing your end-user,
+ which can help OpenAI to monitor and detect abuse.
+ (default: :obj:`None`)
+ tools (list[FunctionTool], optional): A list of tools the model may
+ call. Currently, only functions are supported as a tool. Use this
+ to provide a list of functions the model may generate JSON inputs
+ for. A max of 128 functions are supported.
+ """
+
+ temperature: Optional[float] = None
+ top_p: Optional[float] = None
+ n: Optional[int] = None
+ stream: Optional[bool] = None
+ stop: Optional[Union[str, Sequence[str]]] = None
+ max_tokens: Optional[int] = None
+ presence_penalty: Optional[float] = None
+ response_format: Optional[Union[Type[BaseModel], Dict]] = None
+ frequency_penalty: Optional[float] = None
+ logit_bias: Optional[Dict] = None
+ user: Optional[str] = None
+
+
+NOVITA_API_PARAMS = {param for param in NovitaConfig.model_fields.keys()}
diff --git a/camel/models/__init__.py b/camel/models/__init__.py
index 6e3a2b1bb..443ee3e52 100644
--- a/camel/models/__init__.py
+++ b/camel/models/__init__.py
@@ -32,6 +32,7 @@ from .modelscope_model import ModelScopeModel
from .moonshot_model import MoonshotModel
from .nemotron_model import NemotronModel
from .netmind_model import NetmindModel
+from .novita_model import NovitaModel
from .nvidia_model import NvidiaModel
from .ollama_model import OllamaModel
from .openai_audio_models import OpenAIAudioModels
@@ -68,6 +69,7 @@ __all__ = [
'OpenAIAudioModels',
'NetmindModel',
'NemotronModel',
+ 'NovitaModel',
'NvidiaModel',
'OllamaModel',
'VLLMModel',
diff --git a/camel/models/model_factory.py b/camel/models/model_factory.py
index fac2836d3..342e9a9ec 100644
--- a/camel/models/model_factory.py
+++ b/camel/models/model_factory.py
@@ -32,6 +32,7 @@ from camel.models.mistral_model import MistralModel
from camel.models.modelscope_model import ModelScopeModel
from camel.models.moonshot_model import MoonshotModel
from camel.models.netmind_model import NetmindModel
+from camel.models.novita_model import NovitaModel
from camel.models.nvidia_model import NvidiaModel
from camel.models.ollama_model import OllamaModel
from camel.models.openai_compatible_model import OpenAICompatibleModel
@@ -163,6 +164,8 @@ class ModelFactory:
model_class = MoonshotModel
elif model_platform.is_modelscope:
model_class = ModelScopeModel
+ elif model_platform.is_novita:
+ model_class = NovitaModel
elif model_type == ModelType.STUB:
model_class = StubModel
diff --git a/camel/models/novita_model.py b/camel/models/novita_model.py
new file mode 100644
index 000000000..17099373b
--- /dev/null
+++ b/camel/models/novita_model.py
@@ -0,0 +1,95 @@
+# ========= 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 os
+from typing import Any, Dict, Optional, Union
+
+from camel.configs import NOVITA_API_PARAMS, NovitaConfig
+from camel.models.openai_compatible_model import OpenAICompatibleModel
+from camel.types import ModelType
+from camel.utils import (
+ BaseTokenCounter,
+ api_keys_required,
+)
+
+
+class NovitaModel(OpenAICompatibleModel):
+ r"""Constructor for Novita backend with OpenAI compatibility.
+
+ Args:
+ model_type (Union[ModelType, str]): Model for which a backend is
+ created, supported model can be found here:
+ https://novita.ai/models?utm_source=github_owl&utm_campaign=github_link
+ model_config_dict (Optional[Dict[str, Any]], optional): A dictionary
+ that will be fed into:obj:`openai.ChatCompletion.create()`. If
+ :obj:`None`, :obj:`NovitaConfig().as_dict()` will be used.
+ (default: :obj:`None`)
+ api_key (Optional[str], optional): The API key for authenticating with
+ the Novita service. (default: :obj:`None`)
+ url (Optional[str], optional): The url to the Novita service.
+ If not provided, "https://api.novita.ai/v3/openai" will be used.
+ (default: :obj:`None`)
+ token_counter (Optional[BaseTokenCounter], optional): Token counter to
+ use for the model. If not provided, :obj:`OpenAITokenCounter(
+ ModelType.GPT_4O_MINI)` will be used.
+ timeout (Optional[float], optional): The timeout value in seconds for
+ API calls. If not provided, will fall back to the MODEL_TIMEOUT
+ environment variable or default to 180 seconds.
+ (default: :obj:`None`)
+ """
+
+ @api_keys_required(
+ [
+ ("api_key", 'NOVITA_API_KEY'),
+ ]
+ )
+ def __init__(
+ self,
+ model_type: Union[ModelType, str],
+ model_config_dict: Optional[Dict[str, Any]] = None,
+ api_key: Optional[str] = None,
+ url: Optional[str] = None,
+ token_counter: Optional[BaseTokenCounter] = None,
+ timeout: Optional[float] = None,
+ ) -> None:
+ if model_config_dict is None:
+ model_config_dict = NovitaConfig().as_dict()
+ api_key = api_key or os.environ.get("NOVITA_API_KEY")
+ url = url or os.environ.get(
+ "NOVITA_API_BASE_URL", "https://api.novita.ai/v3/openai"
+ )
+ timeout = timeout or float(os.environ.get("MODEL_TIMEOUT", 180))
+ super().__init__(
+ model_type=model_type,
+ model_config_dict=model_config_dict,
+ api_key=api_key,
+ url=url,
+ token_counter=token_counter,
+ timeout=timeout,
+ )
+
+ def check_model_config(self):
+ r"""Check whether the model configuration contains any
+ unexpected arguments to Novita API.
+
+ Raises:
+ ValueError: If the model configuration dictionary contains any
+ unexpected arguments to Novita API.
+ """
+ for param in self.model_config_dict:
+ if param not in NOVITA_API_PARAMS:
+ raise ValueError(
+ f"Unexpected argument `{param}` is "
+ "input into Novita model backend."
+ )
diff --git a/camel/types/enums.py b/camel/types/enums.py
index f36754d5b..0389db1e3 100644
--- a/camel/types/enums.py
+++ b/camel/types/enums.py
@@ -270,6 +270,63 @@ class ModelType(UnifiedModelType, Enum):
AIML_MIXTRAL_8X7B = "mistralai/Mixtral-8x7B-Instruct-v0.1"
AIML_MISTRAL_7B_INSTRUCT = "mistralai/Mistral-7B-Instruct-v0.1"
+ # Novita platform models support tool calling
+ NOVITA_LLAMA_4_MAVERICK_17B = (
+ "meta-llama/llama-4-maverick-17b-128e-instruct-fp8"
+ )
+ NOVITA_LLAMA_4_SCOUT_17B = "meta-llama/llama-4-scout-17b-16e-instruct"
+ NOVITA_DEEPSEEK_V3_0324 = ("deepseek/deepseek-v3-0324",)
+ NOVITA_QWEN_2_5_V1_72B = ("qwen/qwen2.5-vl-72b-instruct",)
+ NOVITA_DEEPSEEK_V3_TURBO = ("deepseek/deepseek-v3-turbo",)
+ NOVITA_DEEPSEEK_R1_TURBO = ("deepseek/deepseek-r1-turbo",)
+ NOVITA_GEMMA_3_27B_IT = ("google/gemma-3-27b-it",)
+ NOVITA_QWEN_32B = ("qwen/qwq-32b",)
+ NOVITA_L3_8B_STHENO_V3_2 = ("Sao10K/L3-8B-Stheno-v3.2",)
+ NOVITA_MYTHOMAX_L2_13B = ("gryphe/mythomax-l2-13b",)
+ NOVITA_DEEPSEEK_R1_DISTILL_LLAMA_8B = (
+ "deepseek/deepseek-r1-distill-llama-8b",
+ )
+ NOVITA_DEEPSEEK_V3 = ("deepseek/deepseek_v3",)
+ NOVITA_LLAMA_3_1_8B = ("meta-llama/llama-3.1-8b-instruct",)
+ NOVITA_DEEPSEEK_R1_DISTILL_QWEN_14B = (
+ "deepseek/deepseek-r1-distill-qwen-14b",
+ )
+ NOVITA_LLAMA_3_3_70B = ("meta-llama/llama-3.3-70b-instruct",)
+ NOVITA_QWEN_2_5_72B = ("qwen/qwen-2.5-72b-instruct",)
+ NOVITA_MISTRAL_NEMO = ("mistralai/mistral-nemo",)
+ NOVITA_DEEPSEEK_R1_DISTILL_QWEN_32B = (
+ "deepseek/deepseek-r1-distill-qwen-32b",
+ )
+ NOVITA_LLAMA_3_8B = ("meta-llama/llama-3-8b-instruct",)
+ NOVITA_WIZARDLM_2_8X22B = ("microsoft/wizardlm-2-8x22b",)
+ NOVITA_DEEPSEEK_R1_DISTILL_LLAMA_70B = (
+ "deepseek/deepseek-r1-distill-llama-70b",
+ )
+ NOVITA_LLAMA_3_1_70B = ("meta-llama/llama-3.1-70b-instruct",)
+ NOVITA_GEMMA_2_9B_IT = ("google/gemma-2-9b-it",)
+ NOVITA_MISTRAL_7B = ("mistralai/mistral-7b-instruct",)
+ NOVITA_LLAMA_3_70B = ("meta-llama/llama-3-70b-instruct",)
+ NOVITA_DEEPSEEK_R1 = ("deepseek/deepseek-r1",)
+ NOVITA_HERMES_2_PRO_LLAMA_3_8B = ("nousresearch/hermes-2-pro-llama-3-8b",)
+ NOVITA_L3_70B_EURYALE_V2_1 = ("sao10k/l3-70b-euryale-v2.1",)
+ NOVITA_DOLPHIN_MIXTRAL_8X22B = (
+ "cognitivecomputations/dolphin-mixtral-8x22b",
+ )
+ NOVITA_AIROBOROS_L2_70B = ("jondurbin/airoboros-l2-70b",)
+ NOVITA_MIDNIGHT_ROSE_70B = ("sophosympatheia/midnight-rose-70b",)
+ NOVITA_L3_8B_LUNARIS = ("sao10k/l3-8b-lunaris",)
+ NOVITA_GLM_4_9B_0414 = ("thudm/glm-4-9b-0414",)
+ NOVITA_GLM_Z1_9B_0414 = ("thudm/glm-z1-9b-0414",)
+ NOVITA_GLM_Z1_32B_0414 = ("thudm/glm-z1-32b-0414",)
+ NOVITA_GLM_4_32B_0414 = ("thudm/glm-4-32b-0414",)
+ NOVITA_GLM_Z1_RUMINATION_32B_0414 = ("thudm/glm-z1-rumination-32b-0414",)
+ NOVITA_QWEN_2_5_7B = ("qwen/qwen2.5-7b-instruct",)
+ NOVITA_LLAMA_3_2_1B = ("meta-llama/llama-3.2-1b-instruct",)
+ NOVITA_LLAMA_3_2_11B_VISION = ("meta-llama/llama-3.2-11b-vision-instruct",)
+ NOVITA_LLAMA_3_2_3B = ("meta-llama/llama-3.2-3b-instruct",)
+ NOVITA_LLAMA_3_1_8B_BF16 = ("meta-llama/llama-3.1-8b-instruct-bf16",)
+ NOVITA_L31_70B_EURYALE_V2_2 = ("sao10k/l31-70b-euryale-v2.2",)
+
# ModelScope models support tool calling
MODELSCOPE_QWEN_2_5_7B_INSTRUCT = "Qwen/Qwen2.5-7B-Instruct"
MODELSCOPE_QWEN_2_5_14B_INSTRUCT = "Qwen/Qwen2.5-14B-Instruct"
@@ -345,6 +402,7 @@ class ModelType(UnifiedModelType, Enum):
self.is_zhipuai,
self.is_aiml,
self.is_azure_openai,
+ self.is_novita,
]
)
@@ -713,6 +771,54 @@ class ModelType(UnifiedModelType, Enum):
ModelType.SILICONFLOW_PRO_THUDM_GLM_4_9B_CHAT,
}
+ @property
+ def is_novita(self) -> bool:
+ return self in {
+ ModelType.NOVITA_LLAMA_4_MAVERICK_17B,
+ ModelType.NOVITA_LLAMA_4_SCOUT_17B,
+ ModelType.NOVITA_DEEPSEEK_V3_0324,
+ ModelType.NOVITA_QWEN_2_5_V1_72B,
+ ModelType.NOVITA_DEEPSEEK_V3_TURBO,
+ ModelType.NOVITA_DEEPSEEK_R1_TURBO,
+ ModelType.NOVITA_GEMMA_3_27B_IT,
+ ModelType.NOVITA_QWEN_32B,
+ ModelType.NOVITA_L3_8B_STHENO_V3_2,
+ ModelType.NOVITA_MYTHOMAX_L2_13B,
+ ModelType.NOVITA_DEEPSEEK_R1_DISTILL_LLAMA_8B,
+ ModelType.NOVITA_DEEPSEEK_V3,
+ ModelType.NOVITA_LLAMA_3_1_8B,
+ ModelType.NOVITA_DEEPSEEK_R1_DISTILL_QWEN_14B,
+ ModelType.NOVITA_LLAMA_3_3_70B,
+ ModelType.NOVITA_QWEN_2_5_72B,
+ ModelType.NOVITA_MISTRAL_NEMO,
+ ModelType.NOVITA_DEEPSEEK_R1_DISTILL_QWEN_32B,
+ ModelType.NOVITA_LLAMA_3_8B,
+ ModelType.NOVITA_WIZARDLM_2_8X22B,
+ ModelType.NOVITA_DEEPSEEK_R1_DISTILL_LLAMA_70B,
+ ModelType.NOVITA_LLAMA_3_1_70B,
+ ModelType.NOVITA_GEMMA_2_9B_IT,
+ ModelType.NOVITA_MISTRAL_7B,
+ ModelType.NOVITA_LLAMA_3_70B,
+ ModelType.NOVITA_DEEPSEEK_R1,
+ ModelType.NOVITA_HERMES_2_PRO_LLAMA_3_8B,
+ ModelType.NOVITA_L3_70B_EURYALE_V2_1,
+ ModelType.NOVITA_DOLPHIN_MIXTRAL_8X22B,
+ ModelType.NOVITA_AIROBOROS_L2_70B,
+ ModelType.NOVITA_MIDNIGHT_ROSE_70B,
+ ModelType.NOVITA_L3_8B_LUNARIS,
+ ModelType.NOVITA_GLM_4_9B_0414,
+ ModelType.NOVITA_GLM_Z1_9B_0414,
+ ModelType.NOVITA_GLM_Z1_32B_0414,
+ ModelType.NOVITA_GLM_4_32B_0414,
+ ModelType.NOVITA_GLM_Z1_RUMINATION_32B_0414,
+ ModelType.NOVITA_QWEN_2_5_7B,
+ ModelType.NOVITA_LLAMA_3_2_1B,
+ ModelType.NOVITA_LLAMA_3_2_11B_VISION,
+ ModelType.NOVITA_LLAMA_3_2_3B,
+ ModelType.NOVITA_LLAMA_3_1_8B_BF16,
+ ModelType.NOVITA_L31_70B_EURYALE_V2_2,
+ }
+
@property
def is_aiml(self) -> bool:
return self in {
@@ -740,6 +846,9 @@ class ModelType(UnifiedModelType, Enum):
ModelType.COHERE_COMMAND_LIGHT,
ModelType.NVIDIA_NEMOTRON_340B_INSTRUCT,
ModelType.NVIDIA_NEMOTRON_340B_REWARD,
+ ModelType.NOVITA_MYTHOMAX_L2_13B,
+ ModelType.NOVITA_AIROBOROS_L2_70B,
+ ModelType.NOVITA_MIDNIGHT_ROSE_70B,
}:
return 4_096
elif self in {
@@ -761,12 +870,25 @@ class ModelType(UnifiedModelType, Enum):
ModelType.LMSTUDIO_GEMMA_3_4B,
ModelType.LMSTUDIO_GEMMA_3_12B,
ModelType.LMSTUDIO_GEMMA_3_27B,
+ ModelType.NOVITA_L3_8B_STHENO_V3_2,
+ ModelType.NOVITA_LLAMA_3_8B,
+ ModelType.NOVITA_GEMMA_2_9B_IT,
+ ModelType.NOVITA_LLAMA_3_70B,
+ ModelType.NOVITA_HERMES_2_PRO_LLAMA_3_8B,
+ ModelType.NOVITA_L3_70B_EURYALE_V2_1,
+ ModelType.NOVITA_L3_8B_LUNARIS,
+ ModelType.NOVITA_LLAMA_3_1_8B_BF16,
+ ModelType.NOVITA_L31_70B_EURYALE_V2_2,
}:
return 8_192
elif self in {
ModelType.PPIO_BAICHUAN_2_13B_CHAT,
}:
return 14_336
+ elif self in {
+ ModelType.NOVITA_DOLPHIN_MIXTRAL_8X22B,
+ }:
+ return 16_000
elif self in {
ModelType.GPT_3_5_TURBO,
ModelType.YI_LIGHTNING,
@@ -780,11 +902,21 @@ class ModelType(UnifiedModelType, Enum):
ModelType.GLM_4V_PLUS_0111,
ModelType.GLM_ZERO_PREVIEW,
ModelType.PPIO_YI_1_5_34B_CHAT,
+ ModelType.NOVITA_LLAMA_3_1_8B,
}:
return 16_384
-
elif self in {
ModelType.NETMIND_DOUBAO_1_5_PRO,
+ ModelType.NOVITA_GEMMA_3_27B_IT,
+ ModelType.NOVITA_DEEPSEEK_R1_DISTILL_LLAMA_8B,
+ ModelType.NOVITA_QWEN_2_5_72B,
+ ModelType.NOVITA_DEEPSEEK_R1_DISTILL_LLAMA_70B,
+ ModelType.NOVITA_GLM_4_9B_0414,
+ ModelType.NOVITA_GLM_Z1_9B_0414,
+ ModelType.NOVITA_GLM_Z1_32B_0414,
+ ModelType.NOVITA_GLM_4_32B_0414,
+ ModelType.NOVITA_GLM_Z1_RUMINATION_32B_0414,
+ ModelType.NOVITA_QWEN_2_5_7B,
}:
return 32_000
elif self in {
@@ -829,6 +961,11 @@ class ModelType(UnifiedModelType, Enum):
ModelType.MODELSCOPE_MINISTRAL_8B_INSTRUCT,
ModelType.MODELSCOPE_DEEPSEEK_V3_0324,
ModelType.OPENROUTER_LLAMA_3_1_405B,
+ ModelType.NOVITA_QWEN_32B,
+ ModelType.NOVITA_LLAMA_3_1_70B,
+ ModelType.NOVITA_MISTRAL_7B,
+ ModelType.NOVITA_LLAMA_3_2_11B_VISION,
+ ModelType.NOVITA_LLAMA_3_2_3B,
}:
return 32_768
elif self in {
@@ -843,8 +980,22 @@ class ModelType(UnifiedModelType, Enum):
ModelType.PPIO_DEEPSEEK_V3,
ModelType.AWS_DEEPSEEK_R1,
ModelType.NETMIND_QWQ_32B,
+ ModelType.NOVITA_DEEPSEEK_V3_TURBO,
+ ModelType.NOVITA_DEEPSEEK_R1_TURBO,
+ ModelType.NOVITA_DEEPSEEK_V3,
+ ModelType.NOVITA_DEEPSEEK_R1_DISTILL_QWEN_14B,
+ ModelType.NOVITA_DEEPSEEK_R1_DISTILL_QWEN_32B,
+ ModelType.NOVITA_DEEPSEEK_R1,
}:
return 64_000
+ elif self in {
+ ModelType.NOVITA_WIZARDLM_2_8X22B,
+ }:
+ return 65_535
+ elif self in {
+ ModelType.NOVITA_QWEN_2_5_V1_72B,
+ }:
+ return 96_000
elif self in {
ModelType.CLAUDE_2_0,
ModelType.CLAUDE_INSTANT_1_2,
@@ -894,8 +1045,13 @@ class ModelType(UnifiedModelType, Enum):
ModelType.AWS_LLAMA_3_2_11B_INSTRUCT,
ModelType.NETMIND_DEEPSEEK_R1,
ModelType.NETMIND_DEEPSEEK_V3,
+ ModelType.NOVITA_DEEPSEEK_V3_0324,
}:
return 128_000
+ elif self in {
+ ModelType.NOVITA_LLAMA_3_2_1B,
+ }:
+ return 131_000
elif self in {
ModelType.GROQ_LLAMA_3_1_8B,
ModelType.QWEN_PLUS,
@@ -911,6 +1067,9 @@ class ModelType(UnifiedModelType, Enum):
ModelType.OPENROUTER_LLAMA_3_1_70B,
ModelType.PPIO_LLAMA_3_3_70B,
ModelType.OPENROUTER_LLAMA_4_SCOUT,
+ ModelType.NOVITA_LLAMA_4_SCOUT_17B,
+ ModelType.NOVITA_LLAMA_3_3_70B,
+ ModelType.NOVITA_MISTRAL_NEMO,
}:
return 131_072
elif self in {
@@ -961,6 +1120,7 @@ class ModelType(UnifiedModelType, Enum):
ModelType.GPT_4_1,
ModelType.GPT_4_1_MINI,
ModelType.GPT_4_1_NANO,
+ ModelType.NOVITA_LLAMA_4_MAVERICK_17B,
}:
return 1_048_576
elif self in {
@@ -1159,6 +1319,7 @@ class ModelPlatformType(Enum):
AIML = "aiml"
VOLCANO = "volcano"
NETMIND = "netmind"
+ NOVITA = "novita"
@classmethod
def from_name(cls, name):
@@ -1324,6 +1485,11 @@ class ModelPlatformType(Enum):
r"""Returns whether this platform is volcano."""
return self is ModelPlatformType.VOLCANO
+ @property
+ def is_novita(self) -> bool:
+ r"""Returns whether this platform is Novita."""
+ return self is ModelPlatformType.NOVITA
+
class AudioModelType(Enum):
TTS_1 = "tts-1"
diff --git a/camel/types/unified_model_type.py b/camel/types/unified_model_type.py
index 180ad534d..ba50595d2 100644
--- a/camel/types/unified_model_type.py
+++ b/camel/types/unified_model_type.py
@@ -153,6 +153,11 @@ class UnifiedModelType(str):
r"""Returns whether this platform is Moonshot model."""
return True
+ @property
+ def is_novita(self) -> bool:
+ r"""Returns whether the model is a Novita served model."""
+ return True
+
@property
def support_native_structured_output(self) -> bool:
r"""Returns whether the model supports native structured output."""
diff --git a/docs/key_modules/models.md b/docs/key_modules/models.md
index bd32c2aac..78b78d2a3 100644
--- a/docs/key_modules/models.md
+++ b/docs/key_modules/models.md
@@ -36,6 +36,7 @@ CAMEL supports a wide range of models, including [OpenAI’s GPT series](https:/
| **vLLM** | [supported models](https://docs.vllm.ai/en/latest/models/supported_models.html) |
| **SGLANG** | [supported models](https://sgl-project.github.io/references/supported_models.html) |
| **NetMind** | [supported models](https://www.netmind.ai/modelsLibrary) |
+| **NOVITA** | [supported models](https://novita.ai/models?utm_source=github_owl&utm_medium=github_readme&utm_campaign=github_link) |
| **NVIDIA** | [supported models](https://docs.api.nvidia.com/nim/reference/llm-apis) |
| **AIML** | [supported models](https://docs.aimlapi.com/api-overview/model-database/text-models) |
| **ModelScope** | [supported models](https://www.modelscope.cn/docs/model-service/API-Inference/intro) |
diff --git a/examples/models/novita_model_example.py b/examples/models/novita_model_example.py
new file mode 100644
index 000000000..56eebe530
--- /dev/null
+++ b/examples/models/novita_model_example.py
@@ -0,0 +1,52 @@
+# ========= 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. =========
+
+from camel.agents import ChatAgent
+from camel.configs import NovitaConfig
+from camel.models import ModelFactory
+from camel.types import ModelPlatformType, ModelType
+
+model = ModelFactory.create(
+ model_platform=ModelPlatformType.NOVITA,
+ model_type=ModelType.NOVITA_DEEPSEEK_R1_TURBO,
+ model_config_dict=NovitaConfig(temperature=0.2).as_dict(),
+)
+
+# Define system message
+sys_msg = "You are a helpful assistant."
+
+# Set agent
+camel_agent = ChatAgent(system_message=sys_msg, model=model)
+
+user_msg = """Say hi to CAMEL AI, one open-source community
+ dedicated to the study of autonomous and communicative agents."""
+
+# Get response information
+response = camel_agent.step(user_msg)
+print(response.msgs[0].content)
+
+'''
+===============================================================================
+Hi CAMEL AI! 👋 It's fantastic to see an open-source community dedicated to
+advancing autonomous and communicative agents.
+
+Your work in fostering collaboration and innovation is pivotal for the future
+of AI, whether in robotics, NLP, or multi-agent systems. By making this
+research accessible, you're empowering developers and researchers worldwide.
+Keep pushing boundaries—your contributions are shaping a smarter, more
+connected AI landscape!
+
+Wishing you continued growth and breakthroughs! 🚀
+===============================================================================
+'''
diff --git a/test/models/test_novita_model.py b/test/models/test_novita_model.py
new file mode 100644
index 000000000..30adbc140
--- /dev/null
+++ b/test/models/test_novita_model.py
@@ -0,0 +1,62 @@
+# ========= 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 re
+
+import pytest
+
+from camel.configs import NovitaConfig
+from camel.models import NovitaModel
+from camel.types import ModelType
+from camel.utils import OpenAITokenCounter
+
+
+ "model_type",
+ [
+ ModelType.NOVITA_DEEPSEEK_R1,
+ ModelType.NOVITA_LLAMA_3_1_70B,
+ ModelType.NOVITA_QWEN_2_5_72B,
+ ],
+)
+def test_novita_model(model_type: ModelType):
+ model = NovitaModel(model_type)
+ assert model.model_type == model_type
+ assert model.model_config_dict == NovitaConfig().as_dict()
+ assert isinstance(model.token_counter, OpenAITokenCounter)
+ assert isinstance(model.model_type.value_for_tiktoken, str)
+ assert isinstance(model.model_type.token_limit, int)
+
+
+def test_novita_model_unexpected_argument():
+ model_type = ModelType.NOVITA_LLAMA_3_1_70B
+ model_config_dict = {"model_path": "novita_llama_3_1_70B"}
+
+ with pytest.raises(
+ ValueError,
+ match=re.escape(
+ (
+ "Unexpected argument `model_path` is "
+ "input into Novita model backend."
+ )
+ ),
+ ):
+ _ = NovitaModel(model_type, model_config_dict)
+
+
+def test_novita_model_stream_property():
+ model = NovitaModel(ModelType.NOVITA_LLAMA_3_1_70B)
+ assert model.stream is False