Agent Dialog Settings

agent dialog settings configure how your ai assistant responds during conversations, balances memory with performance, and personalizes user experiences these settings are critical for tailoring user interaction quality, managing token usage, and preserving context in long sessions dialog settings dialog settings dialog language field language input (e g , auto ) purpose sets the assistant to respond in the user’s language automatically use case keeps responses consistent with the user’s language without manual intervention required login toggle on/off purpose determines whether user login is required to interact with the ai assistant use case enabled for access control and personalized user tracking disabled for open/public agent usage conversation ai model dropdown options like compact purpose specifies the ai model used to generate responses note "compact" is optimized for q\&a and general dialog, with a 1 5x credit ratio personalization enable personalization toggle on/off purpose enables memory of user profiles and preferences personalization ai model dropdown options include light , compact , thinker , extended and sage credit ratio estimated 1x for "light", 1 5x for "compact", 11x for "thinker" and 20x for "extended" and 100x for "sage" response suggestion suggested response type dropdown options include none , dynamic , or fixed purpose defines how the assistant provides recommended replies to users dialog settings summarization summarize previous answer toggle on/off purpose summarizes previous ai responses to save space and maintain clarity summarization ai model dropdown options include light , compact , thinker , extended and sage credit ratio estimated 1x for "light", 1 5x for "compact", 11x for "thinker" and 20x for "extended" and 100x for "sage" keep function tools answer toggle on/off purpose retains answers from tools (e g , api lookups) for context aware follow ups conversation user message to carry over field numeric (e g , 10 ) purpose determines how many prior user messages are retained in memory for continuity tip lower values improve performance; higher values improve continuity max response token field numeric (e g , 700 ) purpose sets the maximum token length for each assistant reply recommendation balance verbosity and performance using this limit how to configure open dialog settings in your pusaka agent dashboard customize language & authentication preferences choose conversation model suitable for your assistant's use case enable/disable personalization depending on your need to tailor responses set summarization and function answer options to manage memory and tool continuity adjust token and memory limits for optimal performance configure suggested response type for better user engagement click save to apply your changes best practices optimize for use case use compact models for support bots, light models for fast lightweight assistants watch token limits exceeding max token or memory settings may cut off responses enable personalization thoughtfully be transparent with users about what data is retained tune carry over wisely strike the right balance between performance and dialog context next steps once your dialog settings are configured, proceed to update agent behavior or profile for a consistent and goal aligned ai assistant experience