> ## Documentation Index
> Fetch the complete documentation index at: https://art.openpipe.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# GSPO (Group Sequence Policy Optimization)

> A stable and efficient RL algorithm for training language models

<Note>
  GSPO is an experimental feature. The API and behavior may change in future releases.
</Note>

## Overview

GSPO was introduced by the Qwen team to train state-of-the-art models including [Qwen3-235B-A22B-Instruct-2507](https://huggingface.co/Qwen/Qwen3-235B-A22B-Instruct-2507). It can improve training stability and efficiency for Mixture-of-Experts (MoE) models, and may have limited or no impact for dense models.

## Key Benefits

* **Stable Training**: Maintains stable training processes and resolves stability challenges in large MoE models
* **Efficient Scaling**: Achieves higher training efficiency and continues improving with increased computational resources
* **Infrastructure-Friendly**: More tolerant of precision discrepancies, eliminating the need for complex strategies like "Routing Replay"

## How It Works

GSPO's core innovation is its **sequence-level optimization objective**. Instead of focusing on individual token likelihoods, GSPO defines importance ratios based on the **sequence likelihood** with length normalization to reduce variance.

The algorithm optimizes:

```
J_GSPO(θ) = E[1/G ∑ᵢ min(sᵢ(θ) Âᵢ, clip(sᵢ(θ), 1-ε, 1+ε) Âᵢ)]
```

Where the importance ratio `sᵢ(θ)` is defined as:

```
sᵢ(θ) = (π_θ(yᵢ|x) / π_θ_old(yᵢ|x))^(1/|yᵢ|)
```

This sequence-level approach makes GSPO more robust to noise and eliminates the need for complex MoE-specific strategies.

## Configuration

GSPO can be configured using the `importance_sampling_level` parameter when training with ART:

```python theme={null}
model.train(
    trajectory_groups,
    _config=art.dev.TrainConfig(
        importance_sampling_level="sequence",
    )
)
```

## Technical Details

For a deeper understanding of GSPO's technical foundations and comparative analysis with other RL algorithms, see the [original research paper](https://qwenlm.github.io/blog/gspo/).

## Limitations

* As an experimental feature, GSPO may have limited compatibility with some model architectures
* Performance characteristics may vary depending on model size and dataset
* API is subject to change in future releases