---
title: seul-preview
canonical_url: "https://www.modelscope.cn/models/beyoru-ai/seul-preview"
md_url: "https://www.modelscope.cn/models/beyoru-ai/seul-preview.md"
repository: beyoru-ai/seul-preview
last_updated: 2026-07-24
license: apache-2.0
pipeline_tag: text-generation
tasks:
  - text-generation
model_type:
  - qwen3_5
architectures:
  - Qwen3_5ForConditionalGeneration
parameters: 9.4B
tensor_type:
  - BF16
library_name:
  - safetensors
  - pytorch
frameworks:
  - pytorch
downloads: 27
stars: 0
tags:
  - agent
  - tool-use
  - reinforcement-learning
  - grpo
  - rlvr
---

# seul-preview

> seul-preview - beyoru-ai 在 ModelScope 开源的模型。Intelligence grows by accumulation, not replacement.

beyoru-ai/seul-preview 是 ModelScope 魔搭社区上的 9.4B 参数text-generation模型，采用 apache-2.0 许可。

- **Repository**: beyoru-ai/seul-preview
- **License**: apache-2.0
- **Tasks**: text-generation
- **Parameters**: 9.4B
- **Tags**: agent, tool-use, reinforcement-learning, grpo, rlvr
- **Downloads**: 27
- **Stars**: 0
- **Last updated**: 2026-07-24

Source: https://www.modelscope.cn/models/beyoru-ai/seul-preview

---

# seul

> *Intelligence grows by accumulation, not replacement.*

**seul** is the model in a new family of agentic language models, designed for long-horizon reasoning and reliable business tool use.

Rather than optimizing only for benchmark performance, seul is trained to maintain context across extended workflows, interact safely with enterprise tools, and improve through reinforcement learning with verifiable outcomes.


## Design Goals

- Long-horizon reasoning
- Reliable enterprise tool use
- Stable multi-turn planning
- Verifiable execution
- Efficient reinforcement learning


## Limittion

- This training process was conducted on only one domain.
- Due to limited GPU availability, the model was trained for only about half of the originally planned training steps, which may have prevented it from reaching its full potential.

## Citation

If you use **seul** in your research or projects, please cite:

```bibtex
@misc{seul2026,
  title        = {seul-preview},
  author       = {beyoru},
  year         = {2026},
  publisher    = {Hugging Face},
  howpublished = {\url{https://huggingface.co/beyoru/seul-preview}}
}