Title: Feature Request: “Native” Mode for Logic Blocks (Zero-Latency / No AI)
What problem does this feature request solve? The current Logic Block is excellent for the user interface and visual flow, but it forces an LLM inference call for every decision, even for simple variable checks (e.g., {{variable}} is not empty).
This adds unnecessary latency (2+ seconds) to workflows that should be instantaneous. We need the visual structure of the Logic Block without the performance penalty of an AI inference.
What is the use case for this feature? I am building a high-speed bot that needs to instantly decide between two paths:
-
Voice Note: Transcribe Audio
-
Text Message: Process Text
This is a simple boolean check. Currently, the Logic Block sends this to an LLM to “decide,” which causes a massive delay before the bot even starts processing. We simply need to route traffic to specific Jump Blocks based on a variable’s value, instantly.
Please describe the functionality of this feature request. Please enable a “Native” or “Fast” mode for the existing Logic Block based on the Context field.
-
Behavior: When the “Context” field is left empty, the block should automatically skip the LLM inference entirely.
-
Execution: In this state, it should strictly evaluate the Condition rules using system-level logic (0ms latency).
-
Result: This allows us to keep the intuitive UI of the Logic Block for routing to Jump Blocks, but prevents the system from wasting time and money calling an AI model when no “reasoning” is required.
Is there anything else we should know? Debug logs confirm that even when the Context field is completely empty, the Logic Block still triggers an inference event, costing both time and money. If the context is empty, the system should assume it is a logic-only operation. To avoid confusion it could be a separate block titlesd “Router”