instructor 1.3.2__tar.gz → 1.3.4__tar.gz

{instructor-1.3.2 → instructor-1.3.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: instructor
-Version: 1.3.2
+Version: 1.3.4
 Summary: structured outputs for llm
 Home-page: https://github.com/jxnl/instructor
 License: MIT
@@ -20,14 +20,18 @@ Provides-Extra: groq
 Provides-Extra: litellm
 Provides-Extra: mistralai
 Provides-Extra: test-docs
+Provides-Extra: vertexai
 Requires-Dist: aiohttp (>=3.9.1,<4.0.0)
-Requires-Dist: anthropic (>=0.26.0,<0.27.0) ; extra == "anthropic" or extra == "test-docs"
+Requires-Dist: anthropic (>=0.27.0,<0.28.0) ; extra == "anthropic" or extra == "test-docs"
 Requires-Dist: cohere (>=5.1.8,<6.0.0) ; extra == "cohere" or extra == "test-docs"
 Requires-Dist: diskcache (>=5.6.3,<6.0.0) ; extra == "test-docs"
 Requires-Dist: docstring-parser (>=0.16,<0.17)
 Requires-Dist: fastapi (>=0.109.2,<0.110.0) ; extra == "test-docs"
+Requires-Dist: google-cloud-aiplatform (>=1.53.0,<2.0.0) ; extra == "vertexai"
 Requires-Dist: google-generativeai (>=0.5.4,<0.6.0) ; extra == "google-generativeai"
 Requires-Dist: groq (>=0.4.2,<0.5.0) ; extra == "groq" or extra == "test-docs"
+Requires-Dist: jiter (>=0.4.1,<0.5.0)
+Requires-Dist: jsonref (>=1.1.0,<2.0.0) ; extra == "vertexai"
 Requires-Dist: litellm (>=1.35.31,<2.0.0) ; extra == "test-docs" or extra == "litellm"
 Requires-Dist: mistralai (>=0.1.8,<0.2.0) ; extra == "test-docs" or extra == "mistralai"
 Requires-Dist: openai (>=1.1.0,<2.0.0)
@@ -52,6 +56,10 @@ Instructor is a Python library that makes it a breeze to work with structured ou
 [![Discord](https://img.shields.io/discord/1192334452110659664?label=discord)](https://discord.gg/bD9YE9JArw)
 [![Downloads](https://img.shields.io/pypi/dm/instructor.svg)](https://pypi.python.org/pypi/instructor)
 
+## Want your logo on our website?
+
+If your company use instructor a lot, we'd love to have your logo on our website! Please fill out [this form](https://q7gjsgfstrp.typeform.com/to/wluQlVVQ)
+
 ## Key Features
 
 - **Response Models**: Specify Pydantic models to define the structure of your LLM outputs
@@ -184,20 +192,53 @@ import instructor
 import google.generativeai as genai
 from pydantic import BaseModel
 
+
 class User(BaseModel):
     name: str
     age: int
 
+
 # genai.configure(api_key=os.environ["API_KEY"]) # alternative API key configuration
 client = instructor.from_gemini(
     client=genai.GenerativeModel(
-        model_name="models/gemini-1.5-flash-latest", # model defaults to "gemini-pro"
+        model_name="models/gemini-1.5-flash-latest",  # model defaults to "gemini-pro"
     ),
     mode=instructor.Mode.GEMINI_JSON,
 )
+```
+
+
+Alternatively, you can [call Gemini from the OpenAI client](https://cloud.google.com/vertex-ai/generative-ai/docs/multimodal/call-gemini-using-openai-library#python).You'll have to setup [`gcloud`](https://cloud.google.com/docs/authentication/provide-credentials-adc#local-dev), get setup on Vertex AI, and install the Google Auth library.
+
+```sh
+pip install google-auth
+```
+
+```python
+import google.auth
+import google.auth.transport.requests
+import instructor
+from openai import OpenAI
+from pydantic import BaseModel
+
+creds, project = google.auth.default()
+auth_req = google.auth.transport.requests.Request()
+creds.refresh(auth_req)
+
+# Pass the Vertex endpoint and authentication to the OpenAI SDK
+PROJECT = 'PROJECT_ID'
+LOCATION = 'LOCATION' # https://cloud.google.com/vertex-ai/generative-ai/docs/learn/locations
+base_url = f'https://{LOCATION}-aiplatform.googleapis.com/v1beta1/projects/{PROJECT}/locations/{LOCATION}/endpoints/openapi'
+
+client = instructor.from_openai(OpenAI(base_url=base_url, api_key=creds.token), mode=instructor.Mode.JSON)
+# JSON mode is req'd
+class User(BaseModel):
+    name: str
+    age: int
 
-# note that client.chat.completions.create will also work
 resp = client.chat.completions.create(
+    model="google/gemini-1.5-flash-001",
+    max_tokens=1024,
     messages=[
         {
             "role": "user",
@@ -212,6 +253,7 @@ assert resp.name == "Jason"
 assert resp.age == 25
 ```
 
+
 ### Using Litellm
 
 ```python
@@ -369,13 +411,13 @@ for user in user_stream:
     #> name=None age=None
     #> name=None age=None
     #> name=None age=None
-    #> name=None age=25
-    #> name=None age=25
-    #> name=None age=25
-    #> name=None age=25
-    #> name=None age=25
-    #> name=None age=25
-    #> name='John Doe' age=25
+    #> name=None age=None
+    #> name=None age=None
+    #> name='John Doe' age=None
+    #> name='John Doe' age=None
+    #> name='John Doe' age=None
+    #> name='John Doe' age=30
+    #> name='John Doe' age=30
     # name=None age=None
     # name='' age=None
     # name='John' age=None
@@ -415,8 +457,8 @@ users = client.chat.completions.create_iterable(
 
 for user in users:
     print(user)
-    #> name='John' age=30
-    #> name='Jane' age=25
+    #> name='John Doe' age=30
+    #> name='Jane Doe' age=28
     # User(name='John Doe', age=30)
     # User(name='Jane Smith', age=25)
 ```

{instructor-1.3.2 → instructor-1.3.4}/README.md

@@ -6,6 +6,10 @@ Instructor is a Python library that makes it a breeze to work with structured ou
 [![Discord](https://img.shields.io/discord/1192334452110659664?label=discord)](https://discord.gg/bD9YE9JArw)
 [![Downloads](https://img.shields.io/pypi/dm/instructor.svg)](https://pypi.python.org/pypi/instructor)
 
+## Want your logo on our website?
+
+If your company use instructor a lot, we'd love to have your logo on our website! Please fill out [this form](https://q7gjsgfstrp.typeform.com/to/wluQlVVQ)
+
 ## Key Features
 
 - **Response Models**: Specify Pydantic models to define the structure of your LLM outputs
@@ -138,20 +142,53 @@ import instructor
 import google.generativeai as genai
 from pydantic import BaseModel
 
+
 class User(BaseModel):
     name: str
     age: int
 
+
 # genai.configure(api_key=os.environ["API_KEY"]) # alternative API key configuration
 client = instructor.from_gemini(
     client=genai.GenerativeModel(
-        model_name="models/gemini-1.5-flash-latest", # model defaults to "gemini-pro"
+        model_name="models/gemini-1.5-flash-latest",  # model defaults to "gemini-pro"
     ),
     mode=instructor.Mode.GEMINI_JSON,
 )
+```
+
+
+Alternatively, you can [call Gemini from the OpenAI client](https://cloud.google.com/vertex-ai/generative-ai/docs/multimodal/call-gemini-using-openai-library#python).You'll have to setup [`gcloud`](https://cloud.google.com/docs/authentication/provide-credentials-adc#local-dev), get setup on Vertex AI, and install the Google Auth library.
+
+```sh
+pip install google-auth
+```
+
+```python
+import google.auth
+import google.auth.transport.requests
+import instructor
+from openai import OpenAI
+from pydantic import BaseModel
+
+creds, project = google.auth.default()
+auth_req = google.auth.transport.requests.Request()
+creds.refresh(auth_req)
+
+# Pass the Vertex endpoint and authentication to the OpenAI SDK
+PROJECT = 'PROJECT_ID'
+LOCATION = 'LOCATION' # https://cloud.google.com/vertex-ai/generative-ai/docs/learn/locations
+base_url = f'https://{LOCATION}-aiplatform.googleapis.com/v1beta1/projects/{PROJECT}/locations/{LOCATION}/endpoints/openapi'
+
+client = instructor.from_openai(OpenAI(base_url=base_url, api_key=creds.token), mode=instructor.Mode.JSON)
+# JSON mode is req'd
+class User(BaseModel):
+    name: str
+    age: int
 
-# note that client.chat.completions.create will also work
 resp = client.chat.completions.create(
+    model="google/gemini-1.5-flash-001",
+    max_tokens=1024,
     messages=[
         {
             "role": "user",
@@ -166,6 +203,7 @@ assert resp.name == "Jason"
 assert resp.age == 25
 ```
 
+
 ### Using Litellm
 
 ```python
@@ -323,13 +361,13 @@ for user in user_stream:
     #> name=None age=None
     #> name=None age=None
     #> name=None age=None
-    #> name=None age=25
-    #> name=None age=25
-    #> name=None age=25
-    #> name=None age=25
-    #> name=None age=25
-    #> name=None age=25
-    #> name='John Doe' age=25
+    #> name=None age=None
+    #> name=None age=None
+    #> name='John Doe' age=None
+    #> name='John Doe' age=None
+    #> name='John Doe' age=None
+    #> name='John Doe' age=30
+    #> name='John Doe' age=30
     # name=None age=None
     # name='' age=None
     # name='John' age=None
@@ -369,8 +407,8 @@ users = client.chat.completions.create_iterable(
 
 for user in users:
     print(user)
-    #> name='John' age=30
-    #> name='Jane' age=25
+    #> name='John Doe' age=30
+    #> name='Jane Doe' age=28
     # User(name='John Doe', age=30)
     # User(name='Jane Smith', age=25)
 ```

{instructor-1.3.2 → instructor-1.3.4}/instructor/__init__.py

@@ -74,3 +74,8 @@ if importlib.util.find_spec("cohere") is not None:
     from .client_cohere import from_cohere
 
     __all__ += ["from_cohere"]
+
+if importlib.util.find_spec("vertexai") is not None:
+    from .client_vertexai import from_vertexai
+
+    __all__ += ["from_vertexai"]

instructor-1.3.4/instructor/batch.py

@@ -0,0 +1,74 @@
+from typing import Literal, Any, Union
+from pydantic import BaseModel, Field
+from instructor.process_response import handle_response_model
+import uuid
+
+openai_models = Literal[
+    "gpt-4o",
+    "gpt-4-turbo",
+    "gpt-4",
+    "gpt-4-32k",
+    "gpt-3.5-turbo",
+    "gpt-3.5-turbo-16k",
+    "gpt-4-turbo-preview",
+    "gpt-4-vision-preview",
+    "gpt-4-turbo-2024-04-09",
+    "gpt-4-0314",
+    "gpt-4-32k-0314",
+    "gpt-4-32k-0613",
+    "gpt-3.5-turbo-0301",
+    "gpt-3.5-turbo-16k-0613",
+    "gpt-3.5-turbo-1106",
+    "gpt-3.5-turbo-0613",
+]
+
+
+class Function(BaseModel):
+    name: str
+    description: str
+    parameters: Any
+
+
+class Tool(BaseModel):
+    type: str
+    function: Function
+
+
+class RequestBody(BaseModel):
+    model: Union[openai_models, str]
+    messages: list[dict[str, Any]]
+    max_tokens: int = Field(default=1000)
+    tools: list[Tool]
+
+
+class BatchModel(BaseModel):
+    custom_id: str
+    method: Literal["POST"]
+    url: Literal["/v1/chat/completions"]
+    body: RequestBody
+
+
+class BatchJob:
+    @classmethod
+    def create_from_messages(
+        cls,
+        messages_batch: list[list[dict[str, Any]]],
+        model: Union[openai_models, str],
+        response_model: type[BaseModel],
+        max_tokens: int = 1000,
+    ):
+        _, tools = handle_response_model(response_model=response_model)
+        return [
+            BatchModel(
+                custom_id=str(uuid.uuid4()),
+                method="POST",
+                url="/v1/chat/completions",
+                body=RequestBody(
+                    model=model,
+                    max_tokens=max_tokens,
+                    messages=messages,
+                    **tools,
+                ),
+            ).model_dump(mode="json")
+            for messages in messages_batch
+        ]

instructor-1.3.4/instructor/cli/batch.py

@@ -0,0 +1,101 @@
+from rich.console import Console
+from rich.table import Table
+from rich.live import Live
+from openai import OpenAI
+from openai.types.batch import Batch
+import typer
+import datetime
+import time
+
+client = OpenAI()
+app = typer.Typer()
+
+console = Console()
+
+
+def generate_table(batch_jobs: list[Batch]):
+    table = Table(
+        title="OpenAI Batch Jobs",
+    )
+
+    table.add_column("Batch ID", style="dim")
+    table.add_column("Created At")
+    table.add_column("Status")
+    table.add_column("Failed")
+    table.add_column("Completed")
+    table.add_column("Total")
+
+    for batch_job in batch_jobs:
+        table.add_row(
+            batch_job.id,
+            str(datetime.datetime.fromtimestamp(batch_job.created_at)),
+            batch_job.status,
+            str(batch_job.request_counts.failed),  # type: ignore
+            str(batch_job.request_counts.completed),  # type: ignore
+            str(batch_job.request_counts.total),  # type: ignore
+        )
+
+    return table
+
+
+def get_jobs(limit: int = 10):
+    return client.batches.list(limit=limit).data
+
+
+@app.command(name="list", help="See all existing batch jobs")
+def watch(
+    limit: int = typer.Option(10, help="Total number of batch jobs to show"),
+    poll: int = typer.Option(
+        10, help="Time in seconds to wait for the batch job to complete"
+    ),
+    screen: bool = typer.Option(False, help="Enable or disable screen output"),
+):
+    """
+    Monitor the status of the most recent batch jobs
+    """
+    batch_jobs = get_jobs(limit)
+    table = generate_table(batch_jobs)
+    with Live(
+        generate_table(batch_jobs), refresh_per_second=2, screen=screen
+    ) as live_table:
+        while True:
+            batch_jobs = get_jobs(limit)
+            table = generate_table(batch_jobs)
+            live_table.update(table)
+            time.sleep(poll)
+
+
+@app.command(
+    help="Create a batch job from a file",
+)
+def create_from_file(
+    file_path: str = typer.Option(help="File containing the batch job requests"),
+):
+    with console.status(f"[bold green] Uploading batch job file...", spinner="dots"):
+        batch_input_file = client.files.create(
+            file=open(file_path, "rb"), purpose="batch"
+        )
+
+    batch_input_file_id = batch_input_file.id
+
+    with console.status(
+        f"[bold green] Creating batch job from ID {batch_input_file_id}", spinner="dots"
+    ):
+        client.batches.create(
+            input_file_id=batch_input_file_id,
+            endpoint="/v1/chat/completions",
+            completion_window="24h",
+            metadata={"description": "testing job"},
+        )
+
+    watch(limit=5, poll=2, screen=False)
+
+
+@app.command(help="Cancel a batch job")
+def cancel(batch_id: str = typer.Option(help="Batch job ID to cancel")):
+    try:
+        client.batches.cancel(batch_id)
+        watch(limit=5, poll=2, screen=False)
+        console.log(f"[bold red]Job {batch_id} cancelled successfully!")
+    except Exception as e:
+        console.log(f"[bold red]Error cancelling job {batch_id}: {e}")

{instructor-1.3.2 → instructor-1.3.4}/instructor/cli/cli.py

@@ -3,6 +3,7 @@ import instructor.cli.jobs as jobs
 import instructor.cli.files as files
 import instructor.cli.usage as usage
 import instructor.cli.hub as hub
+import instructor.cli.batch as batch
 
 app = typer.Typer()
 
@@ -10,6 +11,7 @@ app.add_typer(jobs.app, name="jobs", help="Monitor and create fine tuning jobs")
 app.add_typer(files.app, name="files", help="Manage files on OpenAI's servers")
 app.add_typer(usage.app, name="usage", help="Check OpenAI API usage data")
 app.add_typer(hub.app, name="hub", help="Interact with the instructor hub")
+app.add_typer(batch.app, name="batch", help="Manage OpenAI Batch jobs")
 
 
 @app.command()
@@ -18,6 +20,6 @@ def docs(query: str = typer.Argument(None, help="Search the documentation")) ->
     Open the instructor documentation website.
     """
     if query:
-        typer.launch(f"https://jxnl.github.io/instructor/?q={query}")
+        typer.launch(f"https://python.useinstructor.com/?q={query}")
     else:
-        typer.launch("https://jxnl.github.io/instructor")
+        typer.launch("https://python.useinstructor.com/")

{instructor-1.3.2 → instructor-1.3.4}/instructor/cli/usage.py

@@ -118,11 +118,11 @@ def calculate_cost(
 
 def group_and_sum_by_date_and_snapshot(usage_data: list[dict[str, Any]]) -> Table:
     """Group and sum the usage data by date and snapshot, including costs."""
-    summary: defaultdict[
-        str, defaultdict[str, dict[str, Union[int, float]]]
-    ] = defaultdict(
-        lambda: defaultdict(
-            lambda: {"total_requests": 0, "total_tokens": 0, "total_cost": 0.0}
+    summary: defaultdict[str, defaultdict[str, dict[str, Union[int, float]]]] = (
+        defaultdict(
+            lambda: defaultdict(
+                lambda: {"total_requests": 0, "total_tokens": 0, "total_cost": 0.0}
+            )
         )
     )
 

{instructor-1.3.2 → instructor-1.3.4}/instructor/client.py

@@ -63,8 +63,7 @@ class Instructor:
         validation_context: dict[str, Any] | None = None,
         strict: bool = True,
         **kwargs: Any,
-    ) -> Awaitable[T]:
-        ...
+    ) -> Awaitable[T]: ...
 
     @overload
     def create(
@@ -75,8 +74,7 @@ class Instructor:
         validation_context: dict[str, Any] | None = None,
         strict: bool = True,
         **kwargs: Any,
-    ) -> T:
-        ...
+    ) -> T: ...
 
     # TODO: we should overload a case where response_model is None
     def create(
@@ -108,8 +106,7 @@ class Instructor:
         validation_context: dict[str, Any] | None = None,
         strict: bool = True,
         **kwargs: Any,
-    ) -> AsyncGenerator[T, None]:
-        ...
+    ) -> AsyncGenerator[T, None]: ...
 
     @overload
     def create_partial(
@@ -120,8 +117,7 @@ class Instructor:
         validation_context: dict[str, Any] | None = None,
         strict: bool = True,
         **kwargs: Any,
-    ) -> Generator[T, None, None]:
-        ...
+    ) -> Generator[T, None, None]: ...
 
     def create_partial(
         self,
@@ -155,8 +151,7 @@ class Instructor:
         validation_context: dict[str, Any] | None = None,
         strict: bool = True,
         **kwargs: Any,
-    ) -> AsyncGenerator[T, None]:
-        ...
+    ) -> AsyncGenerator[T, None]: ...
 
     @overload
     def create_iterable(
@@ -167,8 +162,7 @@ class Instructor:
         validation_context: dict[str, Any] | None = None,
         strict: bool = True,
         **kwargs: Any,
-    ) -> Generator[T, None, None]:
-        ...
+    ) -> Generator[T, None, None]: ...
 
     def create_iterable(
         self,
@@ -179,8 +173,6 @@ class Instructor:
         strict: bool = True,
         **kwargs: Any,
     ) -> Generator[T, None, None] | AsyncGenerator[T, None]:
-        assert self.provider != Provider.ANTHROPIC, "Anthropic doesn't support iterable"
-
         kwargs["stream"] = True
         kwargs = self.handle_kwargs(kwargs)
 
@@ -203,8 +195,7 @@ class Instructor:
         validation_context: dict[str, Any] | None = None,
         strict: bool = True,
         **kwargs: Any,
-    ) -> Awaitable[tuple[T, Any]]:
-        ...
+    ) -> Awaitable[tuple[T, Any]]: ...
 
     @overload
     def create_with_completion(
@@ -215,8 +206,7 @@ class Instructor:
         validation_context: dict[str, Any] | None = None,
         strict: bool = True,
         **kwargs: Any,
-    ) -> tuple[T, Any]:
-        ...
+    ) -> tuple[T, Any]: ...
 
     def create_with_completion(
         self,
@@ -432,8 +422,7 @@ def from_litellm(
     completion: Callable[..., Any],
     mode: instructor.Mode = instructor.Mode.TOOLS,
     **kwargs: Any,
-) -> Instructor:
-    ...
+) -> Instructor: ...
 
 
 @overload

{instructor-1.3.2 → instructor-1.3.4}/instructor/client_anthropic.py

@@ -11,10 +11,9 @@ def from_anthropic(
     client: (
         anthropic.Anthropic | anthropic.AnthropicBedrock | anthropic.AnthropicVertex
     ),
-    mode: instructor.Mode = instructor.Mode.ANTHROPIC_JSON,
+    mode: instructor.Mode = instructor.Mode.ANTHROPIC_TOOLS,
     **kwargs: Any,
-) -> instructor.Instructor:
-    ...
+) -> instructor.Instructor: ...
 
 
 @overload
@@ -24,10 +23,9 @@ def from_anthropic(
         | anthropic.AsyncAnthropicBedrock
         | anthropic.AsyncAnthropicVertex
     ),
-    mode: instructor.Mode = instructor.Mode.ANTHROPIC_JSON,
+    mode: instructor.Mode = instructor.Mode.ANTHROPIC_TOOLS,
     **kwargs: Any,
-) -> instructor.AsyncInstructor:
-    ...
+) -> instructor.AsyncInstructor: ...
 
 
 def from_anthropic(
@@ -39,7 +37,7 @@ def from_anthropic(
         | anthropic.AsyncAnthropicVertex
         | anthropic.AnthropicVertex
     ),
-    mode: instructor.Mode = instructor.Mode.ANTHROPIC_JSON,
+    mode: instructor.Mode = instructor.Mode.ANTHROPIC_TOOLS,
     **kwargs: Any,
 ) -> instructor.Instructor | instructor.AsyncInstructor:
     assert (
@@ -62,10 +60,7 @@ def from_anthropic(
         ),
     ), "Client must be an instance of {anthropic.Anthropic, anthropic.AsyncAnthropic, anthropic.AnthropicBedrock, anthropic.AsyncAnthropicBedrock,  anthropic.AnthropicVertex, anthropic.AsyncAnthropicVertex}"
 
-    if mode == instructor.Mode.ANTHROPIC_TOOLS:
-        create = client.beta.tools.messages.create  # type: ignore - unknown in stubs
-    else:
-        create = client.messages.create
+    create = client.messages.create
 
     if isinstance(
         client,

{instructor-1.3.2 → instructor-1.3.4}/instructor/client_cohere.py

@@ -23,8 +23,7 @@ def from_cohere(
     client: cohere.Client,
     mode: instructor.Mode = instructor.Mode.COHERE_TOOLS,
     **kwargs: Any,
-) -> instructor.Instructor:
-    ...
+) -> instructor.Instructor: ...
 
 
 @overload
@@ -32,8 +31,7 @@ def from_cohere(
     client: cohere.AsyncClient,
     mode: instructor.Mode = instructor.Mode.COHERE_TOOLS,
     **kwargs: Any,
-) -> instructor.AsyncInstructor:
-    ...
+) -> instructor.AsyncInstructor: ...
 
 
 def from_cohere(