nvidia-haystack 0.3.0__tar.gz → 0.5.0__tar.gz

{nvidia_haystack-0.3.0 → nvidia_haystack-0.5.0}/CHANGELOG.md

@@ -1,5 +1,45 @@
 # Changelog
 
+## [integrations/nvidia-v0.4.0] - 2025-10-23
+
+### 🚀 Features
+
+- `NvidiaChatGenerator` add integration tests for mixing Tool/Toolset (#2422)
+
+### 📚 Documentation
+
+- Add pydoc configurations for Docusaurus (#2411)
+- Fix docstrings to avoid errors in API reference generation (#2423)
+
+### 🧪 Testing
+
+- Make tests successfully run from forks (#2203)
+- Remove deprecated NV-Embed-QA model from Nvidia tests (#2370)
+
+### ⚙️ CI
+
+- Temporarily install `click<8.3.0` (#2288)
+
+### 🧹 Chores
+
+- Remove black (#1985)
+- Standardize readmes - part 2 (#2205)
+
+### 🌀 Miscellaneous
+
+- Add structured output support in `NvidiaChatGenerator` (#2405)
+
+## [integrations/nvidia-v0.3.0] - 2025-06-20
+
+### 🐛 Bug Fixes
+
+- Fix Nvidia types + add py.typed (#1970)
+
+### 🧹 Chores
+
+- Align core-integrations Hatch scripts (#1898)
+- Update md files for new hatch scripts (#1911)
+
 ## [integrations/nvidia-v0.2.0] - 2025-06-05
 
 ### 🚀 Features

{nvidia_haystack-0.3.0 → nvidia_haystack-0.5.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: nvidia-haystack
-Version: 0.3.0
+Version: 0.5.0
 Project-URL: Documentation, https://github.com/deepset-ai/haystack-core-integrations/tree/main/integrations/nvidia#readme
 Project-URL: Issues, https://github.com/deepset-ai/haystack-core-integrations/issues
 Project-URL: Source, https://github.com/deepset-ai/haystack-core-integrations/tree/main/integrations/nvidia
@@ -18,7 +18,7 @@ Classifier: Programming Language :: Python :: 3.13
 Classifier: Programming Language :: Python :: Implementation :: CPython
 Classifier: Programming Language :: Python :: Implementation :: PyPy
 Requires-Python: >=3.9
-Requires-Dist: haystack-ai>=2.13.0
+Requires-Dist: haystack-ai>=2.19.0
 Requires-Dist: requests>=2.25.0
 Requires-Dist: tqdm>=4.21.0
 Description-Content-Type: text/markdown
@@ -28,56 +28,18 @@ Description-Content-Type: text/markdown
 [![PyPI - Version](https://img.shields.io/pypi/v/nvidia-haystack.svg)](https://pypi.org/project/nvidia-haystack)
 [![PyPI - Python Version](https://img.shields.io/pypi/pyversions/nvidia-haystack.svg)](https://pypi.org/project/nvidia-haystack)
 
----
-
-**Table of Contents**
-
-- [nvidia-haystack](#nvidia-haystack)
-  - [Installation](#installation)
-  - [Contributing](#contributing)
-  - [License](#license)
-
-## Installation
+- [Integration page](https://haystack.deepset.ai/integrations/nvidia)
+- [Changelog](https://github.com/deepset-ai/haystack-core-integrations/blob/main/integrations/nvidia/CHANGELOG.md)
 
-```console
-pip install nvidia-haystack
-```
+---
 
 ## Contributing
 
-`hatch` is the best way to interact with this project, to install it:
-
-```sh
-pip install hatch
-```
-
-With `hatch` installed, to run all the tests:
-
-```
-hatch run test:all
-```
-
-> Note: integration tests will be skipped unless the env var NVIDIA_API_KEY is set. The api key needs to be valid
-> in order to pass the tests.
-
-To only run unit tests:
-
-```
-hatch run test:unit
-```
-
-To format your code and perform linting using Ruff (with automatic fixes), run:
-
-```
-hatch run fmt
-```
-
-To check for static type errors, run:
-
-```console
-$ hatch run test:types
-```
-
-## License
+Refer to the general [Contribution Guidelines](https://github.com/deepset-ai/haystack-core-integrations/blob/main/CONTRIBUTING.md).
 
-`nvidia-haystack` is distributed under the terms of the [Apache-2.0](https://spdx.org/licenses/Apache-2.0.html) license.
+To run integration tests locally, you need to export the `NVIDIA_API_KEY` environment variable. Some tests require additional environment variables:
+- `NVIDIA_NIM_EMBEDDER_MODEL`
+- `NVIDIA_NIM_ENDPOINT_URL`
+- `NVIDIA_NIM_GENERATOR_MODEL`
+- `NVIDIA_NIM_RANKER_MODEL`
+- `NVIDIA_NIM_RANKER_ENDPOINT_URL`

nvidia_haystack-0.5.0/README.md

@@ -0,0 +1,20 @@
+# nvidia-haystack
+
+[![PyPI - Version](https://img.shields.io/pypi/v/nvidia-haystack.svg)](https://pypi.org/project/nvidia-haystack)
+[![PyPI - Python Version](https://img.shields.io/pypi/pyversions/nvidia-haystack.svg)](https://pypi.org/project/nvidia-haystack)
+
+- [Integration page](https://haystack.deepset.ai/integrations/nvidia)
+- [Changelog](https://github.com/deepset-ai/haystack-core-integrations/blob/main/integrations/nvidia/CHANGELOG.md)
+
+---
+
+## Contributing
+
+Refer to the general [Contribution Guidelines](https://github.com/deepset-ai/haystack-core-integrations/blob/main/CONTRIBUTING.md).
+
+To run integration tests locally, you need to export the `NVIDIA_API_KEY` environment variable. Some tests require additional environment variables:
+- `NVIDIA_NIM_EMBEDDER_MODEL`
+- `NVIDIA_NIM_ENDPOINT_URL`
+- `NVIDIA_NIM_GENERATOR_MODEL`
+- `NVIDIA_NIM_RANKER_MODEL`
+- `NVIDIA_NIM_RANKER_ENDPOINT_URL`

nvidia_haystack-0.5.0/examples/chat_generator_with_structured_outputs.py

@@ -0,0 +1,34 @@
+# SPDX-FileCopyrightText: 2024-present deepset GmbH <info@deepset.ai>
+#
+# SPDX-License-Identifier: Apache-2.0
+
+
+# This example demonstrates how to use the NvidiaChatGenerator component
+# with structured outputs.
+# To run this example, you will need to
+# set `NVIDIA_API_KEY` environment variable
+
+from haystack.dataclasses import ChatMessage
+
+from haystack_integrations.components.generators.nvidia import NvidiaChatGenerator
+
+json_schema = {
+    "type": "object",
+    "properties": {"title": {"type": "string"}, "rating": {"type": "number"}},
+    "required": ["title", "rating"],
+}
+chat_messages = [
+    ChatMessage.from_user(
+        """
+    Return the title and the rating based on the following movie review according
+    to the provided json schema.
+    Review: Inception is a really well made film. I rate it four stars out of five."""
+    )
+]
+
+component = NvidiaChatGenerator(
+    model="meta/llama-3.1-70b-instruct",
+    generation_kwargs={"extra_body": {"nvext": {"guided_json": json_schema}}},
+)
+results = component.run(chat_messages)
+# print(results)

nvidia_haystack-0.5.0/pydoc/config_docusaurus.yml

@@ -0,0 +1,34 @@
+loaders:
+- ignore_when_discovered:
+  - __init__
+  modules:
+  - haystack_integrations.components.embedders.nvidia.document_embedder
+  - haystack_integrations.components.embedders.nvidia.text_embedder
+  - haystack_integrations.components.embedders.nvidia.truncate
+  - haystack_integrations.components.generators.nvidia.generator
+  - haystack_integrations.components.generators.nvidia.chat.chat_generator
+  - haystack_integrations.components.rankers.nvidia.ranker
+  - haystack_integrations.components.rankers.nvidia.truncate
+  search_path:
+  - ../src
+  type: haystack_pydoc_tools.loaders.CustomPythonLoader
+processors:
+- do_not_filter_modules: false
+  documented_only: true
+  expression: null
+  skip_empty_modules: true
+  type: filter
+- type: smart
+- type: crossref
+renderer:
+  description: Nvidia integration for Haystack
+  id: integrations-nvidia
+  markdown:
+    add_member_class_prefix: false
+    add_method_class_prefix: true
+    classdef_code_block: false
+    descriptive_class_title: false
+    descriptive_module_title: true
+    filename: nvidia.md
+  title: Nvidia
+  type: haystack_pydoc_tools.renderers.DocusaurusRenderer

{nvidia_haystack-0.3.0 → nvidia_haystack-0.5.0}/pyproject.toml

@@ -23,7 +23,7 @@ classifiers = [
   "Programming Language :: Python :: Implementation :: CPython",
   "Programming Language :: Python :: Implementation :: PyPy",
 ]
-dependencies = ["haystack-ai>=2.13.0", "requests>=2.25.0", "tqdm>=4.21.0"]
+dependencies = ["haystack-ai>=2.19.0", "requests>=2.25.0", "tqdm>=4.21.0"]
 
 [project.urls]
 Documentation = "https://github.com/deepset-ai/haystack-core-integrations/tree/main/integrations/nvidia#readme"
@@ -46,8 +46,8 @@ installer = "uv"
 dependencies = ["haystack-pydoc-tools", "ruff"]
 
 [tool.hatch.envs.default.scripts]
-docs = ["pydoc-markdown pydoc/config.yml"]
-fmt = "ruff check --fix {args} && ruff format {args}"
+docs = ["pydoc-markdown pydoc/config_docusaurus.yml"]
+fmt = "ruff check --fix {args}; ruff format {args}"
 fmt-check = "ruff check {args} && ruff format --check {args}"
 
 [tool.hatch.envs.test]
@@ -66,7 +66,7 @@ dependencies = [
 unit = 'pytest -m "not integration" {args:tests}'
 integration = 'pytest -m "integration" {args:tests}'
 all = 'pytest {args:tests}'
-cov-retry = 'all --cov=haystack_integrations --reruns 3 --reruns-delay 30 -x'
+cov-retry = 'pytest --cov=haystack_integrations --reruns 3 --reruns-delay 30 -x {args:tests}'
 
 types = """mypy -p haystack_integrations.components.embedders.nvidia \
 -p haystack_integrations.components.generators.nvidia \
@@ -80,13 +80,9 @@ check_untyped_defs = true
 disallow_incomplete_defs = true
 
 
-[tool.black]
-target-version = ["py38"]
-line-length = 120
-skip-string-normalization = true
 
 [tool.ruff]
-target-version = "py38"
+target-version = "py39"
 line-length = 120
 
 [tool.ruff.lint]
@@ -164,4 +160,4 @@ addopts = "--strict-markers"
 markers = [
   "integration: integration tests",
 ]
-log_cli = true
+log_cli = true

{nvidia_haystack-0.3.0 → nvidia_haystack-0.5.0}/src/haystack_integrations/components/embedders/nvidia/document_embedder.py

@@ -4,7 +4,8 @@
 
 import os
 import warnings
-from typing import Any, Dict, List, Optional, Tuple, Union
+from dataclasses import replace
+from typing import Any, Optional, Union
 
 from haystack import Document, component, default_from_dict, default_to_dict, logging
 from haystack.utils import Secret, deserialize_secrets_inplace
@@ -28,7 +29,7 @@ class NvidiaDocumentEmbedder:
 
     doc = Document(content="I love pizza!")
 
-    text_embedder = NvidiaDocumentEmbedder(model="NV-Embed-QA", api_url="https://ai.api.nvidia.com/v1/retrieval/nvidia")
+    text_embedder = NvidiaDocumentEmbedder(model="nvidia/nv-embedqa-e5-v5", api_url="https://integrate.api.nvidia.com/v1")
     text_embedder.warm_up()
 
     result = document_embedder.run([doc])
@@ -45,11 +46,11 @@ class NvidiaDocumentEmbedder:
         suffix: str = "",
         batch_size: int = 32,
         progress_bar: bool = True,
-        meta_fields_to_embed: Optional[List[str]] = None,
+        meta_fields_to_embed: Optional[list[str]] = None,
         embedding_separator: str = "\n",
         truncate: Optional[Union[EmbeddingTruncateMode, str]] = None,
         timeout: Optional[float] = None,
-    ):
+    ) -> None:
         """
         Create a NvidiaTextEmbedder component.
 
@@ -61,7 +62,7 @@ class NvidiaDocumentEmbedder:
             API key for the NVIDIA NIM.
         :param api_url:
             Custom API URL for the NVIDIA NIM.
-            Format for API URL is http://host:port
+            Format for API URL is `http://host:port`
         :param prefix:
             A string to add to the beginning of each text.
         :param suffix:
@@ -108,7 +109,7 @@ class NvidiaDocumentEmbedder:
     def class_name(cls) -> str:
         return "NvidiaDocumentEmbedder"
 
-    def default_model(self):
+    def default_model(self) -> None:
         """Set default model in local NIM mode."""
         valid_models = [
             model.id for model in self.available_models if not model.base_model or model.base_model == model.id
@@ -129,7 +130,7 @@ class NvidiaDocumentEmbedder:
             error_message = "No locally hosted model was found."
             raise ValueError(error_message)
 
-    def warm_up(self):
+    def warm_up(self) -> None:
         """
         Initializes the component.
         """
@@ -156,7 +157,7 @@ class NvidiaDocumentEmbedder:
         if not self.model:
             self.default_model()
 
-    def to_dict(self) -> Dict[str, Any]:
+    def to_dict(self) -> dict[str, Any]:
         """
         Serializes the component to a dictionary.
 
@@ -179,14 +180,14 @@ class NvidiaDocumentEmbedder:
         )
 
     @property
-    def available_models(self) -> List[Model]:
+    def available_models(self) -> list[Model]:
         """
         Get a list of available models that work with NvidiaDocumentEmbedder.
         """
         return self.backend.models() if self.backend else []
 
     @classmethod
-    def from_dict(cls, data: Dict[str, Any]) -> "NvidiaDocumentEmbedder":
+    def from_dict(cls, data: dict[str, Any]) -> "NvidiaDocumentEmbedder":
         """
         Deserializes the component from a dictionary.
 
@@ -200,7 +201,7 @@ class NvidiaDocumentEmbedder:
             deserialize_secrets_inplace(data["init_parameters"], keys=["api_key"])
         return default_from_dict(cls, data)
 
-    def _prepare_texts_to_embed(self, documents: List[Document]) -> List[str]:
+    def _prepare_texts_to_embed(self, documents: list[Document]) -> list[str]:
         texts_to_embed = []
         for doc in documents:
             meta_values_to_embed = [
@@ -213,8 +214,8 @@ class NvidiaDocumentEmbedder:
 
         return texts_to_embed
 
-    def _embed_batch(self, texts_to_embed: List[str], batch_size: int) -> Tuple[List[List[float]], Dict[str, Any]]:
-        all_embeddings: List[List[float]] = []
+    def _embed_batch(self, texts_to_embed: list[str], batch_size: int) -> tuple[list[list[float]], dict[str, Any]]:
+        all_embeddings: list[list[float]] = []
         usage_prompt_tokens = 0
         usage_total_tokens = 0
 
@@ -233,8 +234,8 @@ class NvidiaDocumentEmbedder:
 
         return all_embeddings, {"usage": {"prompt_tokens": usage_prompt_tokens, "total_tokens": usage_total_tokens}}
 
-    @component.output_types(documents=List[Document], meta=Dict[str, Any])
-    def run(self, documents: List[Document]) -> Dict[str, Union[List[Document], Dict[str, Any]]]:
+    @component.output_types(documents=list[Document], meta=dict[str, Any])
+    def run(self, documents: list[Document]) -> dict[str, Union[list[Document], dict[str, Any]]]:
         """
         Embed a list of Documents.
 
@@ -246,14 +247,12 @@ class NvidiaDocumentEmbedder:
             A dictionary with the following keys and values:
             - `documents` - List of processed Documents with embeddings.
             - `meta` - Metadata on usage statistics, etc.
-        :raises RuntimeError:
-            If the component was not initialized.
         :raises TypeError:
-            If the input is not a string.
+            If the input is not a list of Documents.
         """
         if not self._initialized:
-            msg = "The embedding model has not been loaded. Please call warm_up() before running."
-            raise RuntimeError(msg)
+            self.warm_up()
+
         elif not isinstance(documents, list) or (documents and not isinstance(documents[0], Document)):
             msg = (
                 "NvidiaDocumentEmbedder expects a list of Documents as input."
@@ -267,7 +266,9 @@ class NvidiaDocumentEmbedder:
 
         texts_to_embed = self._prepare_texts_to_embed(documents)
         embeddings, metadata = self._embed_batch(texts_to_embed, self.batch_size)
+
+        new_documents = []
         for doc, emb in zip(documents, embeddings):
-            doc.embedding = emb
+            new_documents.append(replace(doc, embedding=emb))
 
-        return {"documents": documents, "meta": metadata}
+        return {"documents": new_documents, "meta": metadata}

{nvidia_haystack-0.3.0 → nvidia_haystack-0.5.0}/src/haystack_integrations/components/embedders/nvidia/text_embedder.py

@@ -4,7 +4,7 @@
 
 import os
 import warnings
-from typing import Any, Dict, List, Optional, Union
+from typing import Any, Optional, Union
 
 from haystack import component, default_from_dict, default_to_dict, logging
 from haystack.utils import Secret, deserialize_secrets_inplace
@@ -30,7 +30,7 @@ class NvidiaTextEmbedder:
 
     text_to_embed = "I love pizza!"
 
-    text_embedder = NvidiaTextEmbedder(model="NV-Embed-QA", api_url="https://ai.api.nvidia.com/v1/retrieval/nvidia")
+    text_embedder = NvidiaTextEmbedder(model="nvidia/nv-embedqa-e5-v5", api_url="https://integrate.api.nvidia.com/v1")
     text_embedder.warm_up()
 
     print(text_embedder.run(text_to_embed))
@@ -58,7 +58,7 @@ class NvidiaTextEmbedder:
             API key for the NVIDIA NIM.
         :param api_url:
             Custom API URL for the NVIDIA NIM.
-            Format for API URL is http://host:port
+            Format for API URL is `http://host:port`
         :param prefix:
             A string to add to the beginning of each text.
         :param suffix:
@@ -146,7 +146,7 @@ class NvidiaTextEmbedder:
             else:
                 self.default_model()
 
-    def to_dict(self) -> Dict[str, Any]:
+    def to_dict(self) -> dict[str, Any]:
         """
         Serializes the component to a dictionary.
 
@@ -165,14 +165,14 @@ class NvidiaTextEmbedder:
         )
 
     @property
-    def available_models(self) -> List[Model]:
+    def available_models(self) -> list[Model]:
         """
         Get a list of available models that work with NvidiaTextEmbedder.
         """
         return self.backend.models() if self.backend else []
 
     @classmethod
-    def from_dict(cls, data: Dict[str, Any]) -> "NvidiaTextEmbedder":
+    def from_dict(cls, data: dict[str, Any]) -> "NvidiaTextEmbedder":
         """
         Deserializes the component from a dictionary.
 
@@ -186,8 +186,8 @@ class NvidiaTextEmbedder:
             deserialize_secrets_inplace(data["init_parameters"], keys=["api_key"])
         return default_from_dict(cls, data)
 
-    @component.output_types(embedding=List[float], meta=Dict[str, Any])
-    def run(self, text: str) -> Dict[str, Union[List[float], Dict[str, Any]]]:
+    @component.output_types(embedding=list[float], meta=dict[str, Any])
+    def run(self, text: str) -> dict[str, Union[list[float], dict[str, Any]]]:
         """
         Embed a string.
 
@@ -197,14 +197,14 @@ class NvidiaTextEmbedder:
             A dictionary with the following keys and values:
             - `embedding` - Embedding of the text.
             - `meta` - Metadata on usage statistics, etc.
-        :raises RuntimeError:
-            If the component was not initialized.
         :raises TypeError:
             If the input is not a string.
+        :raises ValueError:
+            If the input string is empty.
         """
         if not self._initialized:
-            msg = "The embedding model has not been loaded. Please call warm_up() before running."
-            raise RuntimeError(msg)
+            self.warm_up()
+
         elif not isinstance(text, str):
             msg = (
                 "NvidiaTextEmbedder expects a string as an input."

{nvidia_haystack-0.3.0 → nvidia_haystack-0.5.0}/src/haystack_integrations/components/generators/nvidia/chat/chat_generator.py

@@ -3,12 +3,12 @@
 # SPDX-License-Identifier: Apache-2.0
 
 import os
-from typing import Any, Dict, List, Optional, Union
+from typing import Any, Optional
 
 from haystack import component, default_to_dict, logging
 from haystack.components.generators.chat import OpenAIChatGenerator
 from haystack.dataclasses import StreamingCallbackT
-from haystack.tools import Tool, Toolset, serialize_tools_or_toolset
+from haystack.tools import ToolsType, serialize_tools_or_toolset
 from haystack.utils import serialize_callable
 from haystack.utils.auth import Secret
 
@@ -55,12 +55,12 @@ class NvidiaChatGenerator(OpenAIChatGenerator):
         model: str = "meta/llama-3.1-8b-instruct",
         streaming_callback: Optional[StreamingCallbackT] = None,
         api_base_url: Optional[str] = os.getenv("NVIDIA_API_URL", DEFAULT_API_URL),
-        generation_kwargs: Optional[Dict[str, Any]] = None,
-        tools: Optional[Union[List[Tool], Toolset]] = None,
+        generation_kwargs: Optional[dict[str, Any]] = None,
+        tools: Optional[ToolsType] = None,
         timeout: Optional[float] = None,
         max_retries: Optional[int] = None,
-        http_client_kwargs: Optional[Dict[str, Any]] = None,
-    ):
+        http_client_kwargs: Optional[dict[str, Any]] = None,
+    ) -> None:
         """
         Creates an instance of NvidiaChatGenerator.
 
@@ -86,6 +86,22 @@ class NvidiaChatGenerator(OpenAIChatGenerator):
                 comprising the top 10% probability mass are considered.
             - `stream`: Whether to stream back partial progress. If set, tokens will be sent as data-only server-sent
                 events as they become available, with the stream terminated by a data: [DONE] message.
+            - `response_format`: For NVIDIA NIM servers, this parameter has limited support.
+                - The basic JSON mode with `{"type": "json_object"}` is supported by compatible models, to produce
+                valid JSON output.
+                To pass the JSON schema to the model, use the `guided_json` parameter in `extra_body`.
+                For example:
+                ```python
+                generation_kwargs={
+                    "extra_body": {
+                        "nvext": {
+                            "guided_json": {
+                                json_schema
+                        }
+                    }
+                }
+                ```
+                For more details, see the [NVIDIA NIM documentation](https://docs.nvidia.com/nim/large-language-models/latest/structured-generation.html).
         :param tools:
             A list of tools or a Toolset for which the model can prepare calls. This parameter can accept either a
             list of `Tool` objects or a `Toolset` instance.
@@ -110,7 +126,7 @@ class NvidiaChatGenerator(OpenAIChatGenerator):
             http_client_kwargs=http_client_kwargs,
         )
 
-    def to_dict(self) -> Dict[str, Any]:
+    def to_dict(self) -> dict[str, Any]:
         """
         Serialize this component to a dictionary.
 

{nvidia_haystack-0.3.0 → nvidia_haystack-0.5.0}/src/haystack_integrations/components/generators/nvidia/generator.py

@@ -4,7 +4,7 @@
 
 import os
 import warnings
-from typing import Any, Dict, List, Optional, Union
+from typing import Any, Optional, Union
 
 from haystack import component, default_from_dict, default_to_dict
 from haystack.utils.auth import Secret, deserialize_secrets_inplace
@@ -24,7 +24,7 @@ class NvidiaGenerator:
     from haystack_integrations.components.generators.nvidia import NvidiaGenerator
 
     generator = NvidiaGenerator(
-        model="meta/llama3-70b-instruct",
+        model="meta/llama3-8b-instruct",
         model_arguments={
             "temperature": 0.2,
             "top_p": 0.7,
@@ -47,9 +47,9 @@ class NvidiaGenerator:
         model: Optional[str] = None,
         api_url: str = os.getenv("NVIDIA_API_URL", DEFAULT_API_URL),
         api_key: Optional[Secret] = Secret.from_env_var("NVIDIA_API_KEY"),
-        model_arguments: Optional[Dict[str, Any]] = None,
+        model_arguments: Optional[dict[str, Any]] = None,
         timeout: Optional[float] = None,
-    ):
+    ) -> None:
         """
         Create a NvidiaGenerator component.
 
@@ -90,7 +90,7 @@ class NvidiaGenerator:
     def class_name(cls) -> str:
         return "NvidiaGenerator"
 
-    def default_model(self):
+    def default_model(self) -> None:
         """Set default model in local NIM mode."""
         valid_models = [
             model.id for model in self.available_models if not model.base_model or model.base_model == model.id
@@ -111,7 +111,7 @@ class NvidiaGenerator:
             error_message = "No locally hosted model was found."
             raise ValueError(error_message)
 
-    def warm_up(self):
+    def warm_up(self) -> None:
         """
         Initializes the component.
         """
@@ -134,7 +134,7 @@ class NvidiaGenerator:
             else:
                 self.default_model()
 
-    def to_dict(self) -> Dict[str, Any]:
+    def to_dict(self) -> dict[str, Any]:
         """
         Serializes the component to a dictionary.
 
@@ -150,14 +150,14 @@ class NvidiaGenerator:
         )
 
     @property
-    def available_models(self) -> List[Model]:
+    def available_models(self) -> list[Model]:
         """
         Get a list of available models that work with ChatNVIDIA.
         """
         return self.backend.models() if self.backend else []
 
     @classmethod
-    def from_dict(cls, data: Dict[str, Any]) -> "NvidiaGenerator":
+    def from_dict(cls, data: dict[str, Any]) -> "NvidiaGenerator":
         """
         Deserializes the component from a dictionary.
 
@@ -170,8 +170,8 @@ class NvidiaGenerator:
         deserialize_secrets_inplace(init_params, ["api_key"])
         return default_from_dict(cls, data)
 
-    @component.output_types(replies=List[str], meta=List[Dict[str, Any]])
-    def run(self, prompt: str) -> Dict[str, Union[List[str], List[Dict[str, Any]]]]:
+    @component.output_types(replies=list[str], meta=list[dict[str, Any]])
+    def run(self, prompt: str) -> dict[str, Union[list[str], list[dict[str, Any]]]]:
         """
         Queries the model with the provided prompt.
 
@@ -183,8 +183,7 @@ class NvidiaGenerator:
             - `meta` - Metadata for each reply.
         """
         if self.backend is None:
-            msg = "The generation model has not been loaded. Call warm_up() before running."
-            raise RuntimeError(msg)
+            self.warm_up()
 
         assert self.backend is not None
         replies, meta = self.backend.generate(prompt=prompt)