docs: Add missing section headers and fix broken links (#1232)

Co-authored-by: Alex M <alex.martin@cabinetoffice.gov.uk>
Co-authored-by: Jason Liu <jxnl@users.noreply.github.com>
Co-authored-by: devin-ai-integration[bot] <158243242+devin-ai-integration[bot]@users.noreply.github.com>

docs/examples/bulk_classification.md

@@ -528,6 +528,4 @@ async def get_tags(text: List[str], tags: List[Tag]) -> List[Tag]:
 tag_results = asyncio.run(get_tags(text, tags))
 for tag in tag_results:
     print(tag)
-    #> id=0 name='personal'
     #> id=1 name='phone'
-```

docs/integrations/anthropic.md

@@ -15,6 +15,8 @@ pip install "instructor[anthropic]"
 
 Once we've done so, getting started is as simple as using our `from_anthropic` method to patch the client up.
 
+### Basic Usage
+
 ```python
 from pydantic import BaseModel
 from typing import List
@@ -71,6 +73,19 @@ print(user_response.model_dump_json(indent=2))
 """
 ```
 
+### Beta API Features
+
+Anthropic provides beta features through their beta API endpoint. To use these features, enable the beta parameter when creating the client:
+
+```python
+client = instructor.from_anthropic(
+    anthropic.Anthropic(),
+    beta=True  # Enable beta API features
+)
+```
+
+This allows you to access beta features like PDF support and other experimental capabilities. For more information about available beta features, refer to [Anthropic's documentation](https://docs.anthropic.com/claude/docs/beta-features).
+
 ## Streaming Support
 
 Instructor has two main ways that you can use to stream responses out

instructor/client_anthropic.py

@@ -13,6 +13,7 @@ def from_anthropic(
     ),
     mode: instructor.Mode = instructor.Mode.ANTHROPIC_TOOLS,
     enable_prompt_caching: bool = False,
+    beta: bool = False,
     **kwargs: Any,
 ) -> instructor.Instructor: ...
 
@@ -26,6 +27,7 @@ def from_anthropic(
     ),
     mode: instructor.Mode = instructor.Mode.ANTHROPIC_TOOLS,
     enable_prompt_caching: bool = False,
+    beta: bool = False,
     **kwargs: Any,
 ) -> instructor.AsyncInstructor: ...
 
@@ -41,8 +43,25 @@ def from_anthropic(
     ),
     mode: instructor.Mode = instructor.Mode.ANTHROPIC_TOOLS,
     enable_prompt_caching: bool = False,
+    beta: bool = False,
     **kwargs: Any,
 ) -> instructor.Instructor | instructor.AsyncInstructor:
+    """Create an Instructor instance from an Anthropic client.
+
+    Args:
+        client: An instance of Anthropic client (sync or async)
+        mode: The mode to use for the client (ANTHROPIC_JSON or ANTHROPIC_TOOLS)
+        enable_prompt_caching: Whether to enable prompt caching (requires Anthropic or AsyncAnthropic client)
+        beta: Whether to use beta API features (uses client.beta.messages.create)
+        **kwargs: Additional keyword arguments to pass to the Instructor constructor
+
+    Returns:
+        An Instructor instance (sync or async depending on the client type)
+
+    Raises:
+        TypeError: If enable_prompt_caching is True and client is not Anthropic or AsyncAnthropic
+        AssertionError: If mode is not ANTHROPIC_JSON or ANTHROPIC_TOOLS
+    """
     assert (
         mode
         in {
@@ -70,6 +89,8 @@ def from_anthropic(
             raise TypeError(
                 "Client must be an instance of {anthropic.Anthropic, anthropic.AsyncAnthropic} to enable prompt caching"
             )
+    elif beta:
+        create = client.beta.messages.create
     else:
         create = client.messages.create