# Impala implementation in new API stack

**URL:** <https://discuss.ray.io/t/impala-implementation-in-new-api-stack/22962>\
**Category:** RLlib\
**Created:** [August 7, 2025, 8:46pm UTC](https://discuss.ray.io/t/impala-implementation-in-new-api-stack/22962 "2025-08-07T20:46:41Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![waldroje](https://sea2.discourse-cdn.com/flex020/user_avatar/discuss.ray.io/waldroje/32/7527_2.png) [@waldroje](https://discuss.ray.io/u/waldroje)\
**Post date:** [August 7, 2025, 8:46pm UTC](https://discuss.ray.io/t/impala-implementation-in-new-api-stack/22962/1 "2025-08-07T20:46:41Z")

</div>

**1. Severity of the issue: (select one)**  
X None: I’m just curious or want clarification.  
 Low: Annoying but doesn’t hinder my work.  
 Medium: Significantly affects my productivity but can find a workaround.  
 High: Completely blocks me.

**2. Environment:**

- Ray version: 2.45
- Python version: 3.11.10
- OS: linux
- Cloud/Infrastructure: N/A
- Other libs/tools (if relevant):

**3. What happened vs. what you expected:**

- Expected: That there would be a default RLModule for Impala like the other algorithms.
- Actual: There is not.

I am coming back to Ray/Rllib after a long absence, I think I last was engaged just after 2.0… I have old code base of Custom Model\_V2 that I used w/ Impala, and in reading through the docs it would seem I would now have to write a Custom RLModule.. but I see in the algorithms/impala there is no default RLModule, only a Policy and Learner, where the other Algorithms have a RLModule… can you help me out and explain why is that the case?