refactor(image-description): Increase max output tokens to 1024
Previously, the maximum output tokens for image descriptions was set to 256. This was often too restrictive, leading to truncated or incomplete descriptions. Increasing the limit to 1024 allows for more comprehensive and detailed image descriptions while still encouraging conciseness.
This commit is contained in:
parent
1cbacd6f9e
commit
28d0c11e48
@ -2791,7 +2791,7 @@ async def generate_image_description(
|
||||
# System prompt is not strictly needed here as the user prompt is direct.
|
||||
description_gen_config = types.GenerateContentConfig(
|
||||
temperature=0.4, # Lower temperature for more factual description
|
||||
max_output_tokens=256, # Descriptions should be concise
|
||||
max_output_tokens=1024, # Descriptions should be concise
|
||||
safety_settings=STANDARD_SAFETY_SETTINGS,
|
||||
# No response_mime_type or response_schema needed for plain text
|
||||
tools=None, # No tools for this task
|
||||
|
Loading…
x
Reference in New Issue
Block a user