---
license: apache-2.0
base_model: Qwen/Qwen3-32B
pipeline_tag: text-generation
language:
- en
tags:
- text-generation
- reasoning
- thinking
- zen
- zenlm
- hanzo
---

> **Status:** Superseded by [zenlm/zen3-omni](https://huggingface.co/zenlm/zen3-omni). Weights remain available for reproducibility.


# zen-omni-30b-thinking

> Superseded by [zenlm/zen3-omni](https://huggingface.co/zenlm/zen3-omni) — canonical name.

Reasoning-tuned text model. The weights in this repo are a Qwen3-32B-class dense text model.

Repackaged from [Qwen/Qwen3-32B](https://huggingface.co/Qwen/Qwen3-32B) (apache-2.0, Alibaba Qwen). **Not trained from scratch** — a permissively-licensed redistribution for the OSS-clean Zen model line.

## Specs

| Property | Value |
|----------|-------|
| Parameters | 32B (dense) |
| Architecture | Qwen3 (`Qwen3ForCausalLM`) |
| Modality | text |

## License

`apache-2.0`. Upstream: **Qwen/Qwen3-32B** by Alibaba Qwen (apache-2.0).