Skip to content

Drop in a screenshot and convert it to clean code (HTML/Tailwind/React/Vue)

Python

update gemini flash 3 thinking config and max output tokens

Abi Rajaf59204e1bc16
diff --git a/backend/models/gemini.py b/backend/models/gemini.py
index 726a5eeb..ee82e14d 100644
--- a/backend/models/gemini.py
+++ b/backend/models/gemini.py
@@ -59,6 +59,13 @@ async def stream_gemini_response(
thinking_budget=5000, include_thoughts=True
),
)
+ elif model_name == Llm.GEMINI_3_FLASH_PREVIEW.value:
+ # Gemini 3 Flash uses thinking_level instead of thinking_budget
+ config = types.GenerateContentConfig(
+ temperature=0,
+ max_output_tokens=30000,
+ thinking_config=types.ThinkingConfig(thinking_level="minimal"),
+ )
else:
# TODO: Fix output tokens here
config = types.GenerateContentConfig(