make image gen size configurable

vibe_bot/config.py

@@ -38,6 +38,7 @@ EMBEDDING_ENDPOINT_KEY: str = os.getenv("EMBEDDING_ENDPOINT_KEY", "placeholder")
 CHAT_MODEL: str = os.getenv("CHAT_MODEL", "")
 COMPLETION_MODEL: str = os.getenv("COMPLETION_MODEL", "")
 IMAGE_GEN_MODEL: str = os.getenv("IMAGE_GEN_MODEL", "")
+IMAGE_GEN_SIZE: str = os.getenv("IMAGE_GEN_SIZE", "1024x1024")
 IMAGE_EDIT_MODEL: str = os.getenv("IMAGE_EDIT_MODEL", "")
 EMBEDDING_MODEL: str = os.getenv("EMBEDDING_MODEL", "")
 

vibe_bot/llama_wrapper.py

@@ -295,6 +295,7 @@ def image_generation(
     openai_api_key: str,
     model: str = "gen",
     n: int = 1,
+    size: str = "1024x1024",
 ) -> str:
     """Generate an image using the given prompt.
 
@@ -304,6 +305,7 @@ def image_generation(
         openai_api_key: The API key for authentication.
         model: The model to use for image generation.
         n: Number of images to generate.
+        size: The size of the generated image, e.g. "1024x1024".
 
     Returns:
         The base64 encoded image data. Decode and write to a file.
@@ -318,7 +320,7 @@ def image_generation(
         response = client.images.generate(
             prompt=prompt,
             n=n,
-            size="768x768",
+            size=size,
             model=model,
             timeout=120.0,
         )

vibe_bot/main.py

@@ -25,6 +25,7 @@ from vibe_bot.config import (
     IMAGE_GEN_ENDPOINT,
     IMAGE_GEN_ENDPOINT_KEY,
     IMAGE_GEN_MODEL,
+    IMAGE_GEN_SIZE,
     MAX_COMPLETION_TOKENS,
     TTS_MODEL_PATH,
     TTS_SPEED,
@@ -746,6 +747,7 @@ async def doodlebob(ctx: CommandsContext[Bot], *, message: str) -> None:
         openai_url=IMAGE_GEN_ENDPOINT,
         openai_api_key=IMAGE_GEN_ENDPOINT_KEY,
         model=IMAGE_GEN_MODEL,
+        size=IMAGE_GEN_SIZE,
     )
 
     if not image_b64: