---
title: BiliBot-7B-Q
canonical_url: "https://www.modelscope.cn/models/mlx-community/BiliBot-7B-Q"
md_url: "https://www.modelscope.cn/models/mlx-community/BiliBot-7B-Q.md"
repository: mlx-community/BiliBot-7B-Q
last_updated: 2024-12-19
license: apache-2.0
model_type:
  - qwen2
architectures:
  - Qwen2ForCausalLM
parameters: 1.2B
tensor_type:
  - F16
  - U32
library_name:
  - mlx
  - safetensors
  - pytorch
frameworks:
  - pytorch
language:
  - zh
  - en
inference_backends:
  - "deploy_task text/emb"
  - "lmdeploy 0.9.1"
  - "lmdeploy_turbomind 0.9.1"
  - "sglang 0.5.2"
  - "vllm 0.9.2"
downloads: 1002
stars: 0
tags:
  - mlx
---

# BiliBot-7B-Q

> BiliBot-7B-Q - mlx-community 在 ModelScope 开源的模型。mlx-community/BiliBot-7B-Q

mlx-community/BiliBot-7B-Q 是 ModelScope 魔搭社区上的 1.2B 参数机器学习模型，采用 apache-2.0 许可，可用 deploy_task text/emb、lmdeploy 0.9.1、lmdeploy_turbomind 0.9.1 部署。

- **Repository**: mlx-community/BiliBot-7B-Q
- **License**: apache-2.0
- **Parameters**: 1.2B
- **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**: 1002
- **Stars**: 0
- **Last updated**: 2024-12-19

Source: https://www.modelscope.cn/models/mlx-community/BiliBot-7B-Q

---

# mlx-community/BiliBot-7B-Q

The Model [mlx-community/BiliBot-7B-Q](https://huggingface.co/mlx-community/BiliBot-7B-Q) was converted to MLX format from [Kadins/BiliBot-Qwen2-7B-Q-FT](https://huggingface.co/Kadins/BiliBot-Qwen2-7B-Q-FT) using mlx-lm version **0.13.0**.

## Use with mlx

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

```python
from mlx_lm import load, generate

model, tokenizer = load("mlx-community/BiliBot-7B-Q")

# Template content
template = """
<|im_start|>system
You are a helpful assistant<|im_end|>
<|im_start|>user
你是一位B站老用户，请使用暴躁的语言风格，对以下问题给出简短、机智的回答：
{usr_msg}<|im_end|>
<|im_start|>assistant
"""

while True:
    usr_msg = input("用户: ")  # Get user message from terminal
    if usr_msg.lower() == 'quit()':  # Allows the user to exit the loop
        break

    prompt = template.replace("{usr_msg}", usr_msg)

    time_ckpt = time.time()
    response = generate(
        model,
        tokenizer,
        prompt=prompt,
        temp=0.3,
        max_tokens=500,
        verbose=False
    )

    print("%s: %s (Time %d ms)\n" % ("回答", response, (time.time() - time_ckpt) * 1000))
```