Learn how to configure and optimize AI models in the Canvas feature to improve chatbot performance.
0.2
for factual and consistent responses.0.8
for generating innovative ideas or content.100
for short summaries.500
for in-depth explanations.1
to retry failed actions.2–3
for maintaining context across multiple nodes.Image showing the LLM Configuration tab with Temperature, Max Tokens, and Rewind Level settings.
0.1
200
0
0.9
500
2
0.3
300
3
0.4–0.6
) for responses that are both creative and accurate.0.7
).0.2
).0.5
).