---
title: Qwen3-4B-Thinking-2507-DeepSeek-v3.2-Speciale-Code-Distill-qx86-hi-mlx
canonical_url: "https://www.modelscope.cn/models/nightmedia/Qwen3-4B-Thinking-2507-DeepSeek-v3.2-Speciale-Code-Distill-qx86-hi-mlx"
md_url: "https://www.modelscope.cn/models/nightmedia/Qwen3-4B-Thinking-2507-DeepSeek-v3.2-Speciale-Code-Distill-qx86-hi-mlx.md"
repository: nightmedia/Qwen3-4B-Thinking-2507-DeepSeek-v3.2-Speciale-Code-Distill-qx86-hi-mlx
last_updated: 2025-12-25
license: apache-2.0
pipeline_tag: text-generation
tasks:
  - text-generation
model_type:
  - qwen3
architectures:
  - Qwen3ForCausalLM
base_model:
  - TeichAI/Qwen3-4B-Thinking-2507-DeepSeek-v3.2-Speciale-Code-Distill
base_model_relation: quantized
parameters: 1.1B
tensor_type:
  - BF16
  - U32
library_name:
  - mlx
  - safetensors
  - pytorch
frameworks:
  - pytorch
language:
  - en
inference_backends:
  - "deploy_task text/emb"
  - "lmdeploy_turbomind 0.9.1"
  - "sglang 0.5.2"
  - "vllm 0.9.2"
downloads: 242
stars: 0
tags:
  - text-generation-inference
  - transformers
  - unsloth
  - qwen3
  - mlx
---

# Qwen3-4B-Thinking-2507-DeepSeek-v3.2-Speciale-Code-Distill-qx86-hi-mlx

> Qwen3-4B-Thinking-2507-DeepSeek-v3.2-Speciale-Code-Distill-qx86-hi-mlx - nightmedia 在 ModelScope 开源的模型。Qwen3-4B-Thinking-2507-DeepSeek-v3.2-Speciale-Code-Distill-qx86-hi-mlx

nightmedia/Qwen3-4B-Thinking-2507-DeepSeek-v3.2-Speciale-Code-Distill-qx86-hi-mlx 是 ModelScope 魔搭社区上的 1.1B 参数text-generation模型，采用 apache-2.0 许可，基于 TeichAI/Qwen3-4B-Thinking-2507-DeepSeek-v3.2-Speciale-Code-Distill 构建，可用 deploy_task text/emb、lmdeploy_turbomind 0.9.1、sglang 0.5.2 部署。

- **Repository**: nightmedia/Qwen3-4B-Thinking-2507-DeepSeek-v3.2-Speciale-Code-Distill-qx86-hi-mlx
- **License**: apache-2.0
- **Tasks**: text-generation
- **Parameters**: 1.1B
- **Base model**: TeichAI/Qwen3-4B-Thinking-2507-DeepSeek-v3.2-Speciale-Code-Distill
- **Inference backends**: deploy_task text/emb, lmdeploy_turbomind 0.9.1, sglang 0.5.2, vllm 0.9.2
- **Tags**: text-generation-inference, transformers, unsloth, qwen3, mlx
- **Downloads**: 242
- **Stars**: 0
- **Last updated**: 2025-12-25

Source: https://www.modelscope.cn/models/nightmedia/Qwen3-4B-Thinking-2507-DeepSeek-v3.2-Speciale-Code-Distill-qx86-hi-mlx

---

# Qwen3-4B-Thinking-2507-DeepSeek-v3.2-Speciale-Code-Distill-qx86-hi-mlx

> 0.410,0.476,0.677,0.593,0.360,0.714,0.607


This model [Qwen3-4B-Thinking-2507-DeepSeek-v3.2-Speciale-Code-Distill-qx86-hi-mlx](https://huggingface.co/Qwen3-4B-Thinking-2507-DeepSeek-v3.2-Speciale-Code-Distill-qx86-hi-mlx) was
converted to MLX format from [TeichAI/Qwen3-4B-Thinking-2507-DeepSeek-v3.2-Speciale-Code-Distill](https://huggingface.co/TeichAI/Qwen3-4B-Thinking-2507-DeepSeek-v3.2-Speciale-Code-Distill)
using mlx-lm version **0.30.0**.

## Use with mlx

```bash
pip install mlx-lm
```

```python
from mlx_lm import load, generate

model, tokenizer = load("Qwen3-4B-Thinking-2507-DeepSeek-v3.2-Speciale-Code-Distill-qx86-hi-mlx")

prompt = "hello"

if tokenizer.chat_template is not None:
    messages = [{"role": "user", "content": prompt}]
    prompt = tokenizer.apply_chat_template(
        messages, add_generation_prompt=True, return_dict=False,
    )

response = generate(model, tokenizer, prompt=prompt, verbose=True)
```