feat: migrate from Vertex AI models to Google Gemini models (#14)

- Replace VertexAI model classes with GeminiModel for Google AI integration
- Update exception naming: ModelNotDerivedError → ResponseNotDerivedError
- Replace poetry with uv for dependency management
- Update model references throughout codebase and tests
- Refactor language_models module to use google-genai package
- Update documentation to reflect new model structure
- Remove poetry.lock and add uv.lock for new package manager

BREAKING CHANGES:
- Vertex AI models (VertexAIChatModel, VertexAIGenerativeModel, VertexAITextGenerationModel) removed
- Exception class renamed from ModelNotDerivedError to ResponseNotDerivedError
- Package manager changed from poetry to uv
- Google AI models now use GeminiModel instead of Vertex AI classes

Author: christo-olivier

README.md

@@ -3,14 +3,12 @@
 </p>
 
 # Modelsmith
-### Modelsmith is a Python library that allows you to get structured responses in the form of Pydantic models and Python types from Anthropic, Google Vertex AI, and OpenAI models.
+### Modelsmith is a Python library that allows you to get structured responses in the form of Pydantic models and Python types from Anthropic, Google Gemini, and OpenAI models.
 
 Currently it allows you to use the following classes of model:
-- __AnthropicModel__ (used with Anthropic's set of models such as `claude-3-haiku`, `claude-3-sonnet`, `claude-3-opus` and `claude-3_5-sonnet`)
-- __OpenAIModel__ (used with OpenAI's set of models such as `gpt-3.5-turbo`, `gpt-4` and `gpt-4o`)
-- __VertexAIChatModel__ (used with Google Vertex AI's chat models such as `chat-bison`)
-- __VertexAITextGenerationModel__ (used with Google Vertex AI's text generation models such as `text-bison`)
-- __VertexAIGenerativeModel__ (used with Google Vertex AI's generative models such as `gemini-pro`)
+- __AnthropicModel__ (used with Anthropic's full set of models)
+- __OpenAIModel__ (used with OpenAI's full set of models)
+- __GeminiModel__ (used with Google's full set of Gemini models)
 
 Modelsmith provides a unified interface over all of these. It has been designed to be extensible and can adapt to other models in the future.
 

docs/getting_started.md

@@ -18,7 +18,7 @@ class User(BaseModel):
 
 
 # Create your forge instance
-forge = Forge(model=OpenAIModel("gpt-3.5-turbo"), response_model=User)
+forge = Forge(model=OpenAIModel("gpt-5"), response_model=User)
 
 # Generate a User instance from the prompt
 user = forge.generate("Terry Tate 60. Lives in Irvine, United States.")
@@ -31,7 +31,7 @@ print(user)  # name='Terry Tate' age=60 city='Irvine' country='United States'
 Modelsmith does not restrict you to either Pydantic models or Python types. You can combine them in the same response. Below we extract a list of Pydantic model instances.
 
 ```python
-from modelsmith import Forge, VertexAIGenerativeModel
+from modelsmith import Forge, GeminiModel
 from pydantic import BaseModel, Field
 
 
@@ -42,7 +42,7 @@ class City(BaseModel):
 
 # Pass a list of Pydantic models to the response_model argument.
 forge = Forge(
-    model=VertexAIGenerativeModel("gemini-1.5-pro"),
+    model=GeminiModel("gemini-2.5-pro"),
     response_model=list[City],
 )
 
@@ -53,10 +53,10 @@ print(response)  # [City(city='Irvine', state='CA'), City(city='Dallas', state='
 
 ## Using different model types
 
-Using a different model is as simple as passing the desired model class to the Forge. Taking the example above lets use `text-bison` instead of `gemini-pro`.
+Using a different model is as simple as passing the desired model class to the Forge. Taking the example above lets use `gemini-2.5-flash` instead of `gemini-2.5-pro`.
 
 ```python
-from modelsmith import Forge, VertexAITextGenerationModel  # import the correct class
+from modelsmith import Forge, GeminiModel  # import the correct class
 from pydantic import BaseModel, Field
 
 
@@ -65,9 +65,9 @@ class City(BaseModel):
     state: str = Field(description="2-letter abbreviation of the state")
 
 
-# text-bison instead of gemini-pro
+# gemini-2.5-flash instead of gemini-2.5-pro
 forge = Forge(
-    model=VertexAITextGenerationModel("text-bison"),
+    model=GeminiModel("gemini-2.5-flash"),
     response_model=list[City],
 )
 
@@ -76,7 +76,7 @@ response = forge.generate("I have lived in Irvine, CA and Dallas TX")
 print(response)  # [City(city='Irvine', state='CA'), City(city='Dallas', state='TX')]
 ```
 
-If we want to use an Anthropic model the same applies. Simply select the appropriate model class, specify which Anthropic model to use (in this case `claude-3-haiku-20240307`), and pass it to the `Forge` instance.
+If we want to use an Anthropic model the same applies. Simply select the appropriate model class, specify which Anthropic model to use (in this case `claude-sonnet-4-20250514`), and pass it to the `Forge` instance.
 
 ```python
 from modelsmith import Forge, AnthropicModel  # import the correct class
@@ -88,9 +88,9 @@ class City(BaseModel):
     state: str = Field(description="2-letter abbreviation of the state")
 
 
-# Anthropic's claude-3-haiku-20240307 instead of gemini-pro
+# Anthropic's claude-sonnet-4-20250514 instead of gemini-2.5-flash
 forge = Forge(
-    model=AnthropicModel("claude-3-haiku-20240307"),
+    model=AnthropicModel("claude-sonnet-4-20250514"),
     response_model=list[City],
 )
 
@@ -110,7 +110,7 @@ from modelsmith import Forge, VertexAIGenerativeModel
 
 # Create your forge instance
 forge = Forge(
-    model=VertexAIGenerativeModel("gemini-1.5-flash"), response_model=list[str]
+    model=VertexAIGenerativeModel("gemini-2.5-flash"), response_model=list[str]
 )
 
 # Define examples, using inspect.cleandoc to remove indentation
@@ -177,7 +177,7 @@ The same example above would also work if the `response_model_json` was left out
 ```python
 import inspect
 
-from modelsmith import Forge, VertexAITextGenerationModel
+from modelsmith import Forge, GeminiModel
 
 # Create your custom prompt
 my_prompt = inspect.cleandoc("""
@@ -188,7 +188,7 @@ my_prompt = inspect.cleandoc("""
 
 # Create your forge instance, passing your prompt
 forge = Forge(
-    model=VertexAITextGenerationModel("text-bison"),
+    model=GeminiModel("gemini-2.5-flash"),
     response_model=list,
     prompt=my_prompt,
 )
@@ -224,7 +224,7 @@ You can change this by passing the `max_retries` parameter to the `Forge` class.
 ```python
 # Create your forge instance, setting the number of retries
 forge = Forge(
-    model=VertexAIGenerativeModel("gemini-1.0-pro"), response_model=int, max_retries=2
+    model=GeminiModel("gemini-2.5-pro"), response_model=int, max_retries=2
 )
 ```
 
@@ -234,7 +234,7 @@ Modelsmith looks for JSON output in the LLM response. It uses regular expression
 
 ## Failing silently
 
-Modelsmith will raise a `ModelNotDerivedError` exception if no valid response was obtained. You can change this by passing `False` to the `raise_on_failure` parameter of the `Forge` class.
+Modelsmith will raise a `ResponseNotDerivedError` exception if no valid response was obtained. You can change this by passing `False` to the `raise_on_failure` parameter of the `Forge` class.
 
 This will suppress the exception and return `None` instead.
 

docs/index.md

@@ -4,15 +4,11 @@ Modelsmith is a Python library that allows you to get structured responses in th
 
 Currently it allows you to use the following classes of model:
 
-- __AnthropicModel__ (used with Anthropic's set of models such as `claude-3-haiku`, `claude-3-sonnet`, `claude-3-opus` and `claude-3_5-sonnet`)
+- __AnthropicModel__ (used with Anthropic's full set of models)
 
-- __OpenAIModel__ (used with OpenAI's set of models such as `gpt-3.5-turbo`, `gpt-4` and `gpt-4o`)
+- __OpenAIModel__ (used with OpenAI's full set of models)
 
-- __VertexAIChatModel__ (used with Google Vertex AI's chat models such as `chat-bison`)
-
-- __VertexAITextGenerationModel__ (used with Google Vertex AI's text generation models such as `text-bison`)
-
-- __VertexAIGenerativeModel__ (used with Google Vertex AI's generative models such as `gemini-pro`)
+- __GeminiModel__ (used with Google's full set of Gemini models)
 
 
 Modelsmith provides a unified interface over all of these. It has been designed to be extensible and can adapt to other models in the future.

docs/installation.md

@@ -15,7 +15,11 @@ The `AnthropicModel` class takes an optional `api_key` parameter. If not provide
 
 ## Google Cloud Authentication
 
-Authentication to Google Cloud is done via the Application Default Credentials flow. So make sure you have ADC configured. See [Google's documentation](https://cloud.google.com/docs/authentication/provide-credentials-adc) for more details.
+Authentication to Google Cloud is done via either:
+- Application Default Credentials flow. See [Google's documentation](https://cloud.google.com/vertex-ai/generative-ai/docs/start/quickstart?usertype=adc) for more details.
+- Gemini API Key. See [Google Gemini API docs](https://ai.google.dev/gemini-api/docs/api-key?_gl=1*1ya8hy4*_up*MQ..*_ga*MTA2MDc3MjY2MC4xNzU3NzkwNDQz*_ga_P1DBVKWT6V*czE3NTc3OTgwNTMkbzMkZzAkdDE3NTc3OTgwNTMkajYwJGwwJGgxMDQzMjc4MTU2).
+
+The `GeminiModel` allows you to pass the `vertexai`, `api_key`, `project`, and `location` when you initialize the class instance. If you do not pass this in it will be inferred from the environment variables `GOOGLE_GENAI_USE_VERTEXAI`, `GOOGLE_API_KEY`, `GOOGLE_CLOUD_PROJECT`, and `GOOGLE_CLOUD_LOCATION` as per the documentation.
 
 ## Open AI Authentication
 Authentication to OpenAI is done via the OpenAI flow. See the [OpenAI documentation](https://platform.openai.com/docs/quickstart/step-2-set-up-your-api-key) for more details.