---
title: T-pro-it-1.0-Q4-mlx
canonical_url: "https://www.modelscope.cn/models/mlx-community/T-pro-it-1.0-Q4-mlx"
md_url: "https://www.modelscope.cn/models/mlx-community/T-pro-it-1.0-Q4-mlx.md"
repository: mlx-community/T-pro-it-1.0-Q4-mlx
last_updated: 2024-12-19
model_type:
  - qwen2
architectures:
  - Qwen2ForCausalLM
parameters: 5.1B
tensor_type:
  - F16
  - U32
library_name:
  - mlx
  - safetensors
  - pytorch
frameworks:
  - pytorch
language:
  - ru
inference_backends:
  - "deploy_task text/emb"
  - "lmdeploy 0.9.1"
  - "lmdeploy_turbomind 0.9.1"
  - "sglang 0.5.2"
  - "vllm 0.9.2"
downloads: 990
stars: 0
tags:
  - mlx
---

# T-pro-it-1.0-Q4-mlx

> T-pro-it-1.0-Q4-mlx - mlx-community 在 ModelScope 开源的模型。mlx-community/T-Pro-it-1.0-Q4-mlx

mlx-community/T-pro-it-1.0-Q4-mlx 是 ModelScope 魔搭社区上的 5.1B 参数机器学习模型，可用 deploy_task text/emb、lmdeploy 0.9.1、lmdeploy_turbomind 0.9.1 部署。

- **Repository**: mlx-community/T-pro-it-1.0-Q4-mlx
- **Parameters**: 5.1B
- **Inference backends**: deploy_task text/emb, lmdeploy 0.9.1, lmdeploy_turbomind 0.9.1, sglang 0.5.2, vllm 0.9.2
- **Tags**: mlx
- **Downloads**: 990
- **Stars**: 0
- **Last updated**: 2024-12-19

Source: https://www.modelscope.cn/models/mlx-community/T-pro-it-1.0-Q4-mlx

---

# mlx-community/T-Pro-it-1.0-Q4-mlx

The Model [mlx-community/T-pro-it-1.0-Q4-mlx](https://huggingface.co/mlx-community/T-pro-it-1.0-Q4-mlx) was converted to MLX format from [t-tech/T-pro-it-1.0](https://huggingface.co/t-tech/T-pro-it-1.0) using mlx-lm version **0.19.2**.

All rights pertaining to the model are the exclusive property of [T-tech](https://huggingface.co/t-tech), while the model conversion tool is the property of [MLX Community](https://huggingface.co/mlx-community).

## Use with mlx

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

```python
from mlx_lm import load, generate

model, tokenizer = load("mlx-community/T-pro-it-1.0-Q4-mlx")

prompt="Напиши стих про машинное обучение"

if hasattr(tokenizer, "apply_chat_template") and tokenizer.chat_template is not None:
    messages = [
        {"role": "system", "content": "Ты T-pro, виртуальный ассистент в Т-Технологии. Твоя задача - быть полезным диалоговым ассистентом."},
        {"role": "user", "content": prompt}
    ]
    prompt = tokenizer.apply_chat_template(
        messages, tokenize=False, add_generation_prompt=True
    )

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