Skip to content

deprecate interleaved_rollouts#912

Merged
mikasenghaas merged 2 commits intomainfrom
deprecate-interleaved-rollouts
Feb 15, 2026
Merged

deprecate interleaved_rollouts#912
mikasenghaas merged 2 commits intomainfrom
deprecate-interleaved-rollouts

Conversation

@mikasenghaas
Copy link
Copy Markdown
Member

@mikasenghaas mikasenghaas commented Feb 14, 2026

Description

it has no effect anymore. to get tito use client_type="openai_chat_completions_token"

@snimu im not sure why rlm was using this before? it should usually not be env builder's concern to set this

Type of Change

  • Bug fix (non-breaking change which fixes an issue)
  • New feature (non-breaking change which adds functionality)
  • Breaking change (fix or feature that would cause existing functionality to not work as expected)
  • Documentation update
  • Test improvement

Testing

  • All existing tests pass when running uv run pytest locally.
  • New tests have been added to cover the changes

Checklist

  • My code follows the style guidelines of this project as outlined in AGENTS.md
  • I have performed a self-review of my own code
  • I have commented my code, particularly in hard-to-understand areas
  • I have made corresponding changes to the documentation
  • My changes generate no new warnings
  • Any dependent changes have been merged and published

Additional Notes


Note

Medium Risk
This removes a previously accepted configuration/constructor argument (interleaved_rollouts), which may break downstream callers still passing it. Runtime behavior also changes in RLMEnv.get_model_response for cases where sub-LLM steps are included in trajectories.

Overview
Removes the deprecated interleaved_rollouts configuration path across the env stack. Environment and EnvGroup no longer accept/propagate interleaved_rollouts (setter and constructor wiring removed), and docs/reference.md drops it from the public API.

RLMEnv eliminates its set_interleaved_rollouts override and updates get_model_response to apply the “hide trailing sub-LLM trajectory steps” logic whenever include_sub_llm_in_trajectory is enabled (no longer gated by interleaved_rollouts). Tests are updated accordingly to stop setting/asserting interleaved_rollouts and to reflect the new behavior.

Written by Cursor Bugbot for commit a1cb1a7. This will update automatically on new commits. Configure here.

@mikasenghaas mikasenghaas marked this pull request as ready for review February 14, 2026 13:01
Copy link
Copy Markdown

@cursor cursor Bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Cursor Bugbot has reviewed your changes and found 1 potential issue.

Bugbot Autofix is OFF. To automatically fix reported issues with Cloud Agents, enable Autofix in the Cursor dashboard.

Comment thread verifiers/envs/environment.py
@mikasenghaas mikasenghaas requested a review from snimu February 14, 2026 13:12
Follow-up to 44df268 which deprecated interleaved_rollouts.
Removes the set_interleaved_rollouts(bool) method from API reference.

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
@mikasenghaas mikasenghaas merged commit 786d300 into main Feb 15, 2026
6 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants