---
title: wan2.2-t2v-a14b-mlx
canonical_url: "https://www.modelscope.cn/models/Xorbits/wan2.2-t2v-a14b-mlx"
md_url: "https://www.modelscope.cn/models/Xorbits/wan2.2-t2v-a14b-mlx.md"
repository: Xorbits/wan2.2-t2v-a14b-mlx
chinese_name: "Wan2.2 T2V A14B MLX"
last_updated: 2026-08-20
license: apache-2.0
pipeline_tag: text-to-video-synthesis
tasks:
  - text-to-video-synthesis
model_type:
  - t2v
base_model:
  - Wan-AI/Wan2.2-T2V-A14B
base_model_relation: finetune
parameters: 34.4B
tensor_type:
  - BF16
  - F32
library_name:
  - mlx
  - safetensors
downloads: 180
stars: 0
tags:
  - mlx
  - apple-silicon
  - wan2.2
  - text-to-video
  - video-generation
  - mixture-of-experts
---

# wan2.2-t2v-a14b-mlx

> wan2.2-t2v-a14b-mlx - Xorbits 在 ModelScope 开源的模型。Wan2.2-T2V-A14B — MLX (bf16)

Xorbits/wan2.2-t2v-a14b-mlx 是 ModelScope 魔搭社区上的 34.4B 参数text-to-video-synthesis模型，采用 apache-2.0 许可，基于 Wan-AI/Wan2.2-T2V-A14B 构建。

- **Repository**: Xorbits/wan2.2-t2v-a14b-mlx
- **License**: apache-2.0
- **Tasks**: text-to-video-synthesis
- **Parameters**: 34.4B
- **Base model**: Wan-AI/Wan2.2-T2V-A14B
- **Tags**: mlx, apple-silicon, wan2.2, text-to-video, video-generation, mixture-of-experts
- **Downloads**: 180
- **Stars**: 0
- **Last updated**: 2026-08-20

Source: https://www.modelscope.cn/models/Xorbits/wan2.2-t2v-a14b-mlx

---

# Wan2.2-T2V-A14B — MLX (bf16)

Native **MLX** (Apple Silicon) conversion of [Wan-AI/Wan2.2-T2V-A14B](https://huggingface.co/Wan-AI/Wan2.2-T2V-A14B), packaged as a turnkey, self-contained snapshot for the [SceneWorks](https://huggingface.co/SceneWorks) app.

Wan2.2 A14B is a **high/low-noise mixture-of-experts** text-to-video model — two transformers switched at the noise boundary.

## Contents (self-contained, bf16)
| file | what |
|---|---|
| `high_noise_model.safetensors` | high-noise expert DiT (~28.6 GB) |
| `low_noise_model.safetensors` | low-noise expert DiT (~28.6 GB) |
| `t5_encoder.safetensors` | UMT5-XXL text encoder (~11.4 GB) |
| `vae.safetensors` | Wan z16 VAE |
| `tokenizer.json` | UMT5 tokenizer |
| `config.json` | architecture config |

Quantization (Q4/Q8) is applied **at load** by the engine — these weights are full bf16.

## Provenance
- **Source:** `Wan-AI/Wan2.2-T2V-A14B` (Apache-2.0).
- **Converted with:** the SceneWorks native Rust MLX converter (`mlx-gen-wan`, converter id `wan_t2v_14b`), dtype `bfloat16`, dense (no baked-in quant).
- Lean snapshot — only the files the MLX engine loads.

## License
**Apache-2.0**, inherited from the upstream model. This repository redistributes a converted copy of the upstream Apache-2.0 weights, with attribution, as permitted by that license. See the [source model card](https://huggingface.co/Wan-AI/Wan2.2-T2V-A14B).