feat: add generative-deepseek module

test/collection/test_classes_generative.py

@@ -182,6 +182,31 @@ def test_generative_parameters_images_parsing(
                 ),
             ),
         ),
+        (
+            GenerativeConfig.deepseek(
+                base_url="http://localhost:8080",
+                model="deepseek-chat",
+                temperature=0.5,
+                max_tokens=100,
+                frequency_penalty=0.1,
+                presence_penalty=0.2,
+                top_p=0.9,
+                stop=["\n"],
+            )._to_grpc(_GenerativeConfigRuntimeOptions(return_metadata=True)),
+            generative_pb2.GenerativeProvider(
+                return_metadata=True,
+                deepseek=generative_pb2.GenerativeDeepseek(
+                    base_url="http://localhost:8080",
+                    model="deepseek-chat",
+                    temperature=0.5,
+                    max_tokens=100,
+                    frequency_penalty=0.1,
+                    presence_penalty=0.2,
+                    top_p=0.9,
+                    stop=base_pb2.TextArray(values=["\n"]),
+                ),
+            ),
+        ),
         (
             GenerativeConfig.dummy()._to_grpc(
                 _GenerativeConfigRuntimeOptions(return_metadata=True)

test/collection/test_config.py

@@ -1104,6 +1104,30 @@ def test_config_with_vectorizer_and_properties(
             }
         },
     ),
+    (
+        Configure.Generative.deepseek(
+            model="deepseek-chat",
+            max_tokens=100,
+            temperature=0.5,
+            frequency_penalty=0.1,
+            presence_penalty=0.2,
+            top_p=0.9,
+            base_url="https://api.deepseek.com",
+            stop=["\n"],
+        ),
+        {
+            "generative-deepseek": {
+                "model": "deepseek-chat",
+                "maxTokens": 100,
+                "temperature": 0.5,
+                "frequencyPenalty": 0.1,
+                "presencePenalty": 0.2,
+                "topP": 0.9,
+                "baseURL": "https://api.deepseek.com",
+                "stop": ["\n"],
+            }
+        },
+    ),
     (
         Configure.Generative.xai(
             model="grok-2-latest",

weaviate/collections/classes/config.py

@@ -214,6 +214,7 @@ class GenerativeSearches(str, BaseEnum):
         COHERE: Weaviate module backed by Cohere generative models.
         CONTEXTUALAI: Weaviate module backed by ContextualAI generative models.
         DATABRICKS: Weaviate module backed by Databricks generative models.
+        DEEPSEEK: Weaviate module backed by DeepSeek generative models.
         FRIENDLIAI: Weaviate module backed by FriendliAI generative models.
         MISTRAL: Weaviate module backed by Mistral generative models.
         NVIDIA: Weaviate module backed by NVIDIA generative models.
@@ -228,6 +229,7 @@ class GenerativeSearches(str, BaseEnum):
     COHERE = "generative-cohere"
     CONTEXTUALAI = "generative-contextualai"
     DATABRICKS = "generative-databricks"
+    DEEPSEEK = "generative-deepseek"
     DUMMY = "generative-dummy"
     FRIENDLIAI = "generative-friendliai"
     MISTRAL = "generative-mistral"
@@ -443,6 +445,20 @@ class _GenerativeDatabricks(_GenerativeProvider):
     topP: Optional[float]
 
 
+class _GenerativeDeepseek(_GenerativeProvider):
+    generative: Union[GenerativeSearches, _EnumLikeStr] = Field(
+        default=GenerativeSearches.DEEPSEEK, frozen=True, exclude=True
+    )
+    model: Optional[str]
+    temperature: Optional[float]
+    maxTokens: Optional[int]
+    frequencyPenalty: Optional[float]
+    presencePenalty: Optional[float]
+    topP: Optional[float]
+    baseURL: Optional[str]
+    stop: Optional[List[str]]
+
+
 class _GenerativeMistral(_GenerativeProvider):
     generative: Union[GenerativeSearches, _EnumLikeStr] = Field(
         default=GenerativeSearches.MISTRAL, frozen=True, exclude=True
@@ -753,6 +769,41 @@ def databricks(
             topP=top_p,
         )
 
+    @staticmethod
+    def deepseek(
+        *,
+        base_url: Optional[str] = None,
+        model: Optional[str] = None,
+        temperature: Optional[float] = None,
+        max_tokens: Optional[int] = None,
+        frequency_penalty: Optional[float] = None,
+        presence_penalty: Optional[float] = None,
+        top_p: Optional[float] = None,
+        stop: Optional[List[str]] = None,
+    ) -> _GenerativeProvider:
+        """Create a `_GenerativeDeepseek` object for use when performing AI generation using the `generative-deepseek` module.
+
+        Args:
+            base_url: The base URL where the API request should go. Defaults to `None`, which uses the server-defined default
+            model: The model to use. Defaults to `None`, which uses the server-defined default
+            temperature: The temperature to use. Defaults to `None`, which uses the server-defined default
+            max_tokens: The maximum number of tokens to generate. Defaults to `None`, which uses the server-defined default
+            frequency_penalty: The frequency penalty to use. Defaults to `None`, which uses the server-defined default
+            presence_penalty: The presence penalty to use. Defaults to `None`, which uses the server-defined default
+            top_p: The top P value to use. Defaults to `None`, which uses the server-defined default
+            stop: The stop sequences to use. Defaults to `None`, which uses the server-defined default
+        """
+        return _GenerativeDeepseek(
+            model=model,
+            temperature=temperature,
+            maxTokens=max_tokens,
+            frequencyPenalty=frequency_penalty,
+            presencePenalty=presence_penalty,
+            topP=top_p,
+            baseURL=base_url,
+            stop=stop,
+        )
+
     @staticmethod
     def friendliai(
         *,

weaviate/collections/classes/generative.py

@@ -205,6 +205,36 @@ def _to_grpc(self, opts: _GenerativeConfigRuntimeOptions) -> generative_pb2.Gene
         )
 
 
+class _GenerativeDeepseek(_GenerativeConfigRuntime):
+    generative: Union[GenerativeSearches, _EnumLikeStr] = Field(
+        default=GenerativeSearches.DEEPSEEK, frozen=True, exclude=True
+    )
+    base_url: Optional[AnyHttpUrl]
+    model: Optional[str]
+    temperature: Optional[float]
+    max_tokens: Optional[int]
+    frequency_penalty: Optional[float]
+    presence_penalty: Optional[float]
+    top_p: Optional[float]
+    stop: Optional[List[str]]
+
+    def _to_grpc(self, opts: _GenerativeConfigRuntimeOptions) -> generative_pb2.GenerativeProvider:
+        self._validate_multi_modal(opts)
+        return generative_pb2.GenerativeProvider(
+            return_metadata=opts.return_metadata,
+            deepseek=generative_pb2.GenerativeDeepseek(
+                base_url=_parse_anyhttpurl(self.base_url),
+                model=self.model,
+                temperature=self.temperature,
+                max_tokens=self.max_tokens,
+                frequency_penalty=self.frequency_penalty,
+                presence_penalty=self.presence_penalty,
+                top_p=self.top_p,
+                stop=_to_text_array(self.stop),
+            ),
+        )
+
+
 class _GenerativeDummy(_GenerativeConfigRuntime):
     generative: Union[GenerativeSearches, _EnumLikeStr] = Field(
         default=GenerativeSearches.DUMMY, frozen=True, exclude=True
@@ -793,6 +823,43 @@ def databricks(
             top_p=top_p,
         )
 
+    @staticmethod
+    def deepseek(
+        *,
+        base_url: Optional[str] = None,
+        model: Optional[str] = None,
+        temperature: Optional[float] = None,
+        max_tokens: Optional[int] = None,
+        frequency_penalty: Optional[float] = None,
+        presence_penalty: Optional[float] = None,
+        top_p: Optional[float] = None,
+        stop: Optional[List[str]] = None,
+    ) -> _GenerativeConfigRuntime:
+        """Create a `_GenerativeDeepseek` object for use when performing AI generation using the `generative-deepseek` module.
+
+        Args:
+            base_url: The base URL where the API request should go. Defaults to `None`, which uses the server-defined default
+            model: The model to use. Defaults to `None`, which uses the server-defined default
+            temperature: The temperature to use. Defaults to `None`, which uses the server-defined default
+            max_tokens: The maximum number of tokens to generate. Defaults to `None`, which uses the server-defined default
+            frequency_penalty: The frequency penalty to use. Defaults to `None`, which uses the server-defined default
+            presence_penalty: The presence penalty to use. Defaults to `None`, which uses the server-defined default
+            top_p: The top P value to use. Defaults to `None`, which uses the server-defined default
+            stop: The stop sequences to use. Defaults to `None`, which uses the server-defined default
+        """
+        return _GenerativeDeepseek(
+            base_url=TypeAdapter(AnyHttpUrl).validate_python(base_url)
+            if base_url is not None
+            else None,
+            model=model,
+            temperature=temperature,
+            max_tokens=max_tokens,
+            frequency_penalty=frequency_penalty,
+            presence_penalty=presence_penalty,
+            top_p=top_p,
+            stop=stop,
+        )
+
     @staticmethod
     def dummy() -> _GenerativeConfigRuntime:
         """Create a `_GenerativeDummy` object for use when performing AI generation using the `generative-dummy` module."""

weaviate/proto/v1/v4216/v1/generative_pb2.pyi

@@ -42,7 +42,7 @@ class GenerativeSearch(_message.Message):
     def __init__(self, single_response_prompt: _Optional[str] = ..., grouped_response_task: _Optional[str] = ..., grouped_properties: _Optional[_Iterable[str]] = ..., single: _Optional[_Union[GenerativeSearch.Single, _Mapping]] = ..., grouped: _Optional[_Union[GenerativeSearch.Grouped, _Mapping]] = ...) -> None: ...
 
 class GenerativeProvider(_message.Message):
-    __slots__ = ["return_metadata", "anthropic", "anyscale", "aws", "cohere", "dummy", "mistral", "ollama", "openai", "google", "databricks", "friendliai", "nvidia", "xai", "contextualai"]
+    __slots__ = ["return_metadata", "anthropic", "anyscale", "aws", "cohere", "dummy", "mistral", "ollama", "openai", "google", "databricks", "friendliai", "nvidia", "xai", "contextualai", "deepseek"]
     RETURN_METADATA_FIELD_NUMBER: _ClassVar[int]
     ANTHROPIC_FIELD_NUMBER: _ClassVar[int]
     ANYSCALE_FIELD_NUMBER: _ClassVar[int]
@@ -58,6 +58,7 @@ class GenerativeProvider(_message.Message):
     NVIDIA_FIELD_NUMBER: _ClassVar[int]
     XAI_FIELD_NUMBER: _ClassVar[int]
     CONTEXTUALAI_FIELD_NUMBER: _ClassVar[int]
+    DEEPSEEK_FIELD_NUMBER: _ClassVar[int]
     return_metadata: bool
     anthropic: GenerativeAnthropic
     anyscale: GenerativeAnyscale
@@ -73,7 +74,8 @@ class GenerativeProvider(_message.Message):
     nvidia: GenerativeNvidia
     xai: GenerativeXAI
     contextualai: GenerativeContextualAI
-    def __init__(self, return_metadata: bool = ..., anthropic: _Optional[_Union[GenerativeAnthropic, _Mapping]] = ..., anyscale: _Optional[_Union[GenerativeAnyscale, _Mapping]] = ..., aws: _Optional[_Union[GenerativeAWS, _Mapping]] = ..., cohere: _Optional[_Union[GenerativeCohere, _Mapping]] = ..., dummy: _Optional[_Union[GenerativeDummy, _Mapping]] = ..., mistral: _Optional[_Union[GenerativeMistral, _Mapping]] = ..., ollama: _Optional[_Union[GenerativeOllama, _Mapping]] = ..., openai: _Optional[_Union[GenerativeOpenAI, _Mapping]] = ..., google: _Optional[_Union[GenerativeGoogle, _Mapping]] = ..., databricks: _Optional[_Union[GenerativeDatabricks, _Mapping]] = ..., friendliai: _Optional[_Union[GenerativeFriendliAI, _Mapping]] = ..., nvidia: _Optional[_Union[GenerativeNvidia, _Mapping]] = ..., xai: _Optional[_Union[GenerativeXAI, _Mapping]] = ..., contextualai: _Optional[_Union[GenerativeContextualAI, _Mapping]] = ...) -> None: ...
+    deepseek: GenerativeDeepseek
+    def __init__(self, return_metadata: bool = ..., anthropic: _Optional[_Union[GenerativeAnthropic, _Mapping]] = ..., anyscale: _Optional[_Union[GenerativeAnyscale, _Mapping]] = ..., aws: _Optional[_Union[GenerativeAWS, _Mapping]] = ..., cohere: _Optional[_Union[GenerativeCohere, _Mapping]] = ..., dummy: _Optional[_Union[GenerativeDummy, _Mapping]] = ..., mistral: _Optional[_Union[GenerativeMistral, _Mapping]] = ..., ollama: _Optional[_Union[GenerativeOllama, _Mapping]] = ..., openai: _Optional[_Union[GenerativeOpenAI, _Mapping]] = ..., google: _Optional[_Union[GenerativeGoogle, _Mapping]] = ..., databricks: _Optional[_Union[GenerativeDatabricks, _Mapping]] = ..., friendliai: _Optional[_Union[GenerativeFriendliAI, _Mapping]] = ..., nvidia: _Optional[_Union[GenerativeNvidia, _Mapping]] = ..., xai: _Optional[_Union[GenerativeXAI, _Mapping]] = ..., contextualai: _Optional[_Union[GenerativeContextualAI, _Mapping]] = ..., deepseek: _Optional[_Union[GenerativeDeepseek, _Mapping]] = ...) -> None: ...
 
 class GenerativeAnthropic(_message.Message):
     __slots__ = ["base_url", "max_tokens", "model", "temperature", "top_k", "top_p", "stop_sequences", "images", "image_properties"]
@@ -375,6 +377,26 @@ class GenerativeContextualAI(_message.Message):
     knowledge: _base_pb2.TextArray
     def __init__(self, model: _Optional[str] = ..., temperature: _Optional[float] = ..., top_p: _Optional[float] = ..., max_new_tokens: _Optional[int] = ..., system_prompt: _Optional[str] = ..., avoid_commentary: bool = ..., knowledge: _Optional[_Union[_base_pb2.TextArray, _Mapping]] = ...) -> None: ...
 
+class GenerativeDeepseek(_message.Message):
+    __slots__ = ["base_url", "model", "temperature", "max_tokens", "frequency_penalty", "presence_penalty", "top_p", "stop"]
+    BASE_URL_FIELD_NUMBER: _ClassVar[int]
+    MODEL_FIELD_NUMBER: _ClassVar[int]
+    TEMPERATURE_FIELD_NUMBER: _ClassVar[int]
+    MAX_TOKENS_FIELD_NUMBER: _ClassVar[int]
+    FREQUENCY_PENALTY_FIELD_NUMBER: _ClassVar[int]
+    PRESENCE_PENALTY_FIELD_NUMBER: _ClassVar[int]
+    TOP_P_FIELD_NUMBER: _ClassVar[int]
+    STOP_FIELD_NUMBER: _ClassVar[int]
+    base_url: str
+    model: str
+    temperature: float
+    max_tokens: int
+    frequency_penalty: float
+    presence_penalty: float
+    top_p: float
+    stop: _base_pb2.TextArray
+    def __init__(self, base_url: _Optional[str] = ..., model: _Optional[str] = ..., temperature: _Optional[float] = ..., max_tokens: _Optional[int] = ..., frequency_penalty: _Optional[float] = ..., presence_penalty: _Optional[float] = ..., top_p: _Optional[float] = ..., stop: _Optional[_Union[_base_pb2.TextArray, _Mapping]] = ...) -> None: ...
+
 class GenerativeAnthropicMetadata(_message.Message):
     __slots__ = ["usage"]
     class Usage(_message.Message):
@@ -569,8 +591,23 @@ class GenerativeXAIMetadata(_message.Message):
     usage: GenerativeXAIMetadata.Usage
     def __init__(self, usage: _Optional[_Union[GenerativeXAIMetadata.Usage, _Mapping]] = ...) -> None: ...
 
+class GenerativeDeepseekMetadata(_message.Message):
+    __slots__ = ["usage"]
+    class Usage(_message.Message):
+        __slots__ = ["prompt_tokens", "completion_tokens", "total_tokens"]
+        PROMPT_TOKENS_FIELD_NUMBER: _ClassVar[int]
+        COMPLETION_TOKENS_FIELD_NUMBER: _ClassVar[int]
+        TOTAL_TOKENS_FIELD_NUMBER: _ClassVar[int]
+        prompt_tokens: int
+        completion_tokens: int
+        total_tokens: int
+        def __init__(self, prompt_tokens: _Optional[int] = ..., completion_tokens: _Optional[int] = ..., total_tokens: _Optional[int] = ...) -> None: ...
+    USAGE_FIELD_NUMBER: _ClassVar[int]
+    usage: GenerativeDeepseekMetadata.Usage
+    def __init__(self, usage: _Optional[_Union[GenerativeDeepseekMetadata.Usage, _Mapping]] = ...) -> None: ...
+
 class GenerativeMetadata(_message.Message):
-    __slots__ = ["anthropic", "anyscale", "aws", "cohere", "dummy", "mistral", "ollama", "openai", "google", "databricks", "friendliai", "nvidia", "xai"]
+    __slots__ = ["anthropic", "anyscale", "aws", "cohere", "dummy", "mistral", "ollama", "openai", "google", "databricks", "friendliai", "nvidia", "xai", "deepseek"]
     ANTHROPIC_FIELD_NUMBER: _ClassVar[int]
     ANYSCALE_FIELD_NUMBER: _ClassVar[int]
     AWS_FIELD_NUMBER: _ClassVar[int]
@@ -584,6 +621,7 @@ class GenerativeMetadata(_message.Message):
     FRIENDLIAI_FIELD_NUMBER: _ClassVar[int]
     NVIDIA_FIELD_NUMBER: _ClassVar[int]
     XAI_FIELD_NUMBER: _ClassVar[int]
+    DEEPSEEK_FIELD_NUMBER: _ClassVar[int]
     anthropic: GenerativeAnthropicMetadata
     anyscale: GenerativeAnyscaleMetadata
     aws: GenerativeAWSMetadata
@@ -597,7 +635,8 @@ class GenerativeMetadata(_message.Message):
     friendliai: GenerativeFriendliAIMetadata
     nvidia: GenerativeNvidiaMetadata
     xai: GenerativeXAIMetadata
-    def __init__(self, anthropic: _Optional[_Union[GenerativeAnthropicMetadata, _Mapping]] = ..., anyscale: _Optional[_Union[GenerativeAnyscaleMetadata, _Mapping]] = ..., aws: _Optional[_Union[GenerativeAWSMetadata, _Mapping]] = ..., cohere: _Optional[_Union[GenerativeCohereMetadata, _Mapping]] = ..., dummy: _Optional[_Union[GenerativeDummyMetadata, _Mapping]] = ..., mistral: _Optional[_Union[GenerativeMistralMetadata, _Mapping]] = ..., ollama: _Optional[_Union[GenerativeOllamaMetadata, _Mapping]] = ..., openai: _Optional[_Union[GenerativeOpenAIMetadata, _Mapping]] = ..., google: _Optional[_Union[GenerativeGoogleMetadata, _Mapping]] = ..., databricks: _Optional[_Union[GenerativeDatabricksMetadata, _Mapping]] = ..., friendliai: _Optional[_Union[GenerativeFriendliAIMetadata, _Mapping]] = ..., nvidia: _Optional[_Union[GenerativeNvidiaMetadata, _Mapping]] = ..., xai: _Optional[_Union[GenerativeXAIMetadata, _Mapping]] = ...) -> None: ...
+    deepseek: GenerativeDeepseekMetadata
+    def __init__(self, anthropic: _Optional[_Union[GenerativeAnthropicMetadata, _Mapping]] = ..., anyscale: _Optional[_Union[GenerativeAnyscaleMetadata, _Mapping]] = ..., aws: _Optional[_Union[GenerativeAWSMetadata, _Mapping]] = ..., cohere: _Optional[_Union[GenerativeCohereMetadata, _Mapping]] = ..., dummy: _Optional[_Union[GenerativeDummyMetadata, _Mapping]] = ..., mistral: _Optional[_Union[GenerativeMistralMetadata, _Mapping]] = ..., ollama: _Optional[_Union[GenerativeOllamaMetadata, _Mapping]] = ..., openai: _Optional[_Union[GenerativeOpenAIMetadata, _Mapping]] = ..., google: _Optional[_Union[GenerativeGoogleMetadata, _Mapping]] = ..., databricks: _Optional[_Union[GenerativeDatabricksMetadata, _Mapping]] = ..., friendliai: _Optional[_Union[GenerativeFriendliAIMetadata, _Mapping]] = ..., nvidia: _Optional[_Union[GenerativeNvidiaMetadata, _Mapping]] = ..., xai: _Optional[_Union[GenerativeXAIMetadata, _Mapping]] = ..., deepseek: _Optional[_Union[GenerativeDeepseekMetadata, _Mapping]] = ...) -> None: ...
 
 class GenerativeReply(_message.Message):
     __slots__ = ["result", "debug", "metadata"]