# Condition node or Filter node?

**URL:** <https://community.sparkflows.ai/t/condition-node-or-filter-node/403>\
**Category:** Agentic AI\
**Tags:** faq, user-guide, genai\
**Created:** [November 11, 2026, 2:21am UTC](https://community.sparkflows.ai/t/condition-node-or-filter-node/403 "2026-11-11T02:21:19Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![Aryan](https://avatars.discourse-cdn.com/v4/letter/a/d78d45/32.png) [@Aryan](https://community.sparkflows.ai/u/Aryan)\
**Post date:** [November 11, 2026, 2:21am UTC](https://community.sparkflows.ai/t/condition-node-or-filter-node/403/1 "2026-11-11T02:21:19Z")

</div>

- **Condition** decides where the run goes: one rule, two outlets ( **T** and **F** ). It reads the values of earlier nodes by name.
- **Filter** works on a list: it keeps the records that match and drops the rest. Its rule reads one record at a time as `item`, for example `item["amount"] > 100`.

Both share the Conditions builder. A misspelled field in a Filter fails the step and names the available fields instead of quietly passing zero rows.

A Filter can also judge records with the model ( **AI evaluation** with a criterion in plain words) when the rule cannot be written as a formula.
