Add example (#448)

* fix spelling mistake

* update to claude4

* update readme

* add example video

---------

Co-authored-by: Your Name <you@example.com>

Author: dillonalaird

README.md

@@ -22,7 +22,9 @@ _Prompt with an image/video → Get runnable vision code → Build Visual AI App
 
 <br />
 
-**VisionAgent** is the Visual AI pilot from LandingAI. Give it a prompt and an image, and it automatically picks the right vision models and outputs ready‑to‑run code—letting you build vision‑enabled apps in minutes.
+**VisionAgent** is the Visual AI pilot from LandingAI. Give it a prompt and an image, and it automatically picks the right vision models and outputs ready‑to‑run code—letting you build vision‑enabled apps in minutes. You can play around with VisionAgent using our local webapp in `examples/chat` and following the directions in the `README.md`:
+
+<https://github.com/user-attachments/assets/752632b3-dda5-44f1-b27e-5cb4c97757ac>
 
 
 ## Steps to Set Up the Library  

examples/chat/chat-app/src/components/VersionSelector.tsx

@@ -15,7 +15,7 @@ const versionInfo = {
     description: "Latest version that's faster and better at reasoning but has fewer features.",
     features: [
       "Improved reasoning and planning",
-      "Can hanlde image inputs",
+      "Can handle image inputs",
       "Can do object detection",
       "More stable and faster but less feature-rich",
       "Final output is the final answer, not code",

vision_agent/agent/vision_agent_v3.py

@@ -173,7 +173,7 @@ def __init__(
     ) -> None:
         if agent is None:
             self.agent = AnthropicLMM(
-                model_name="claude-3-7-sonnet-20250219", max_tokens=8192
+                model_name="claude-sonnet-4-20250514", max_tokens=8192
             )
         self.kwargs = {
             "thinking": {"type": "enabled", "budget_tokens": 4096},