unslothai/unsloth

How to set "reasoning_effort" of GPT-OSS during GRPO rollouts?

Closed

#3,949 opened on Jan 29, 2026

View on GitHubContributor guide
 (3 comments) (0 reactions) (0 assignees)Python (5,658 forks)batch import
help wanted

Repository metrics

Stars
 (64,271 stars)
PR merge metrics
 (Avg merge 3d 15h) (525 merged PRs in 30d)

Description

  1. Did you update? Yes
  2. Colab or Kaggle or local / cloud. Kaggle
  3. Number GPUs used, use nvidia-smi. 1
  4. Which trainer? GRPOTrainer

Question: We can see how to set reasoning_effort when manually inference in official colab example:

inputs = tokenizer.apply_chat_template(
    messages,
    add_generation_prompt = True,
    return_tensors = "pt",
    return_dict = True,
    reasoning_effort = "low", # **NEW!** Set reasoning effort to low, medium or high
).to("cuda")

_ = model.generate(**inputs, max_new_tokens = 64, streamer = TextStreamer(tokenizer))

But how to set reasoning_effort of GRPO Trainer (rollouts)? I could not find this option in official colab examples. I have tested that maybeGRPOTrainer is using "high" by default . But for me "medium" is enough and more time-efficient.

Thanks in advance!

Contributor guide

Research direction
Examine the GRPOTrainer's rollout generation code to see if 'reasoning effort' is supported. If not, modify the generation call to pass the parameter from the training configuration.
Tech stack
pythonpytorch
Domain
aibackend
Issue type
Feature
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
2
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
1-3 hours
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
Active
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
Clear
Prerequisites
Python
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
65