Skip to content

Fix RL Training - vLLM sampler TypeError - #5101

Open
chiajunglien wants to merge 1 commit into
mainfrom
emma/fix-rl-0902
Open

Fix RL Training - vLLM sampler TypeError#5101
chiajunglien wants to merge 1 commit into
mainfrom
emma/fix-rl-0902

Conversation

@chiajunglien

Copy link
Copy Markdown
Collaborator

Description

This PR resolves critical failures in RL training pipelines. It restores a previously dropped fix for the MaxTextVllmSampler initialization.

DAG Error Log

Root Cause

  • RL Rollout TypeError: Commit 4521fc5 accidentally overwrote the fix applied in 5bf4b88 due to a stale branch merge. It reintroduced the invalid converter=converter kwarg into the base tunix VllmSampler.__init__, triggering an immediate TypeError during RL graph initialization.

Solution

Rollout Logic: Removed the invalid converter argument from super().__init__ in MaxTextVllmSampler, restoring the logic without breaking the newly added Qwen3.5 routing.

Tests

Llama3.1-70b RL: https://cloudlogging.app.goo.gl/i7cbeR8hzSyt38s36

Checklist

Before submitting this PR, please make sure (put X in square brackets):

  • I have performed a self-review of my code. For an optional AI review, add the gemini-review label.
  • I have necessary comments in my code, particularly in hard-to-understand areas.
  • I have run end-to-end tests tests and provided workload links above if applicable.
  • I have made or will make corresponding changes to the doc if needed, including adding new documentation pages to the relevant Table of Contents (toctree directive) as explained in our documentation.

@gemini-code-assist gemini-code-assist Bot left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

Code Review

This pull request updates the __init__ method in maxtext_vllm_rollout.py by removing the converter argument from the super().__init__ call and removing the self.converter = converter assignment. The reviewer notes that while removing converter from super().__init__ is correct, removing self.converter = converter will break weight conversion because the base class expects this attribute. A suggestion is provided to restore the assignment.

Comment thread src/maxtext/integration/vllm/maxtext_vllm_rollout.py
@chiajunglien chiajunglien changed the title Fix vLLM sampler init failures Fix RL Training - vLLM sampler TypeError Sep 2, 2026
- Remove invalid `converter` kwarg from `MaxTextVllmSampler.__init__` .
@codecov

codecov Bot commented Sep 2, 2026

Copy link
Copy Markdown

Codecov Report

❌ Patch coverage is 0% with 1 line in your changes missing coverage. Please review.

Files with missing lines Patch % Lines
...c/maxtext/integration/vllm/maxtext_vllm_rollout.py 0.00% 1 Missing ⚠️

📢 Thoughts on this report? Let us know!

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.

1 participant