---
title: Qwen3-VL-Reranker-2B-6bit
canonical_url: "https://www.modelscope.cn/models/mlx-community/Qwen3-VL-Reranker-2B-6bit"
md_url: "https://www.modelscope.cn/models/mlx-community/Qwen3-VL-Reranker-2B-6bit.md"
repository: mlx-community/Qwen3-VL-Reranker-2B-6bit
last_updated: 2026-03-14
license: apache-2.0
pipeline_tag: text-ranking
tasks:
  - text-ranking
model_type:
  - qwen3_vl
architectures:
  - Qwen3VLForConditionalGeneration
base_model:
  - Qwen/Qwen3-VL-2B-Instruct
base_model_relation: finetune
parameters: 783.4M
tensor_type:
  - U32
  - F16
library_name:
  - mlx
  - safetensors
  - pytorch
frameworks:
  - pytorch
downloads: 83
stars: 0
tags:
  - transformers
  - "multimodal rerank"
  - "text rerank"
  - mlx
---

# Qwen3-VL-Reranker-2B-6bit

> Qwen3-VL-Reranker-2B-6bit - mlx-community 在 ModelScope 开源的模型。mlx-community/Qwen3-VL-Reranker-2B-6bit

mlx-community/Qwen3-VL-Reranker-2B-6bit 是 ModelScope 魔搭社区上的 783.4M 参数text-ranking模型，采用 apache-2.0 许可，基于 Qwen/Qwen3-VL-2B-Instruct 构建。

- **Repository**: mlx-community/Qwen3-VL-Reranker-2B-6bit
- **License**: apache-2.0
- **Tasks**: text-ranking
- **Parameters**: 783.4M
- **Base model**: Qwen/Qwen3-VL-2B-Instruct
- **Tags**: transformers, multimodal rerank, text rerank, mlx
- **Downloads**: 83
- **Stars**: 0
- **Last updated**: 2026-03-14

Source: https://www.modelscope.cn/models/mlx-community/Qwen3-VL-Reranker-2B-6bit

---

# mlx-community/Qwen3-VL-Reranker-2B-6bit

The Model [mlx-community/Qwen3-VL-Reranker-2B-6bit](https://huggingface.co/mlx-community/Qwen3-VL-Reranker-2B-6bit) was converted to MLX format from [Qwen/Qwen3-VL-Reranker-2B](https://huggingface.co/Qwen/Qwen3-VL-Reranker-2B) using mlx-lm version **0.1.0**.

## Use with mlx

```bash
pip install mlx-embeddings
```

```python
from mlx_embeddings import load, generate
import mlx.core as mx

model, tokenizer = load("mlx-community/Qwen3-VL-Reranker-2B-6bit")

# For image-text embeddings
images = [
    "./images/cats.jpg",  # cats
]
texts = ["a photo of cats", "a photo of a desktop setup", "a photo of a person"]

# Process all image-text pairs
outputs = generate(model, processor, texts, images=images)
logits_per_image = outputs.logits_per_image
probs = mx.sigmoid(logits_per_image) # probabilities for this image
for i, image in enumerate(images):
    print(f"Image {i+1}:")
    for j, text in enumerate(texts):
        print(f"  {probs[i][j]:.1%} match with '{text}'")
    print()


```