---
title: RakutenAI-2.0-8x7B-MLX-8bit
canonical_url: "https://www.modelscope.cn/models/mlx-community/RakutenAI-2.0-8x7B-MLX-8bit"
md_url: "https://www.modelscope.cn/models/mlx-community/RakutenAI-2.0-8x7B-MLX-8bit.md"
repository: mlx-community/RakutenAI-2.0-8x7B-MLX-8bit
last_updated: 2026-03-20
license: apache-2.0
pipeline_tag: text-generation
tasks:
  - text-generation
model_type:
  - mixtral
architectures:
  - MixtralForCausalLM
base_model:
  - Rakuten/RakutenAI-2.0-8x7B
base_model_relation: quantized
parameters: 13.2B
tensor_type:
  - BF16
  - U32
library_name:
  - mlx
  - safetensors
  - pytorch
frameworks:
  - pytorch
language:
  - en
  - ja
inference_backends:
  - "deploy_task text/emb"
  - "lmdeploy 0.9.1"
  - "lmdeploy_turbomind 0.9.1"
  - "sglang 0.5.2"
  - "vllm 0.9.2"
downloads: 10
stars: 0
tags:
  - mlx
---

# RakutenAI-2.0-8x7B-MLX-8bit

> RakutenAI-2.0-8x7B-MLX-8bit - mlx-community 在 ModelScope 开源的模型。This model was converted to MLX format from Rakuten/RakutenAI-2.0-8x7B using mlx-lm version 0.31.0. Refer to the original model card for more details on the model.

mlx-community/RakutenAI-2.0-8x7B-MLX-8bit 是 ModelScope 魔搭社区上的 13.2B 参数text-generation模型，采用 apache-2.0 许可，基于 Rakuten/RakutenAI-2.0-8x7B 构建，可用 deploy_task text/emb、lmdeploy 0.9.1、lmdeploy_turbomind 0.9.1 部署。

- **Repository**: mlx-community/RakutenAI-2.0-8x7B-MLX-8bit
- **License**: apache-2.0
- **Tasks**: text-generation
- **Parameters**: 13.2B
- **Base model**: Rakuten/RakutenAI-2.0-8x7B
- **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**: 10
- **Stars**: 0
- **Last updated**: 2026-03-20

Source: https://www.modelscope.cn/models/mlx-community/RakutenAI-2.0-8x7B-MLX-8bit

---

This model was converted to MLX format from [`Rakuten/RakutenAI-2.0-8x7B`](https://huggingface.co/Rakuten/RakutenAI-2.0-8x7B) using mlx-lm version **0.31.0**.
Refer to the [original model card](https://huggingface.co/Rakuten/RakutenAI-2.0-8x7B) for more details on the model.

## Use with mlx

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

```python
from mlx_lm import load, generate

model, tokenizer = load("mlx-community/RakutenAI-2.0-8x7B-MLX-8bit")

prompt = "hello"

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

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

```
Inference: M3 Ultra
==========
 world - 简书
hello world
hello world
hello world
推荐阅读 更多精彩内容
Android - 收藏集
Android 自定义View的各种姿势1 Activity的显示之ViewRootImpl详解 Activity...
passiontim 阅读 163,326 评论 24 赞 695
Android - 收藏集
用两张图告诉你，为什么你的 App 会卡顿? - Android - 掘金 Cover 有什么料？ 从这篇文章中你...
hw1212 阅读 10,519 评论 2 赞 52
Android - 收藏集
用两张图告诉你，为什么你的 App 会卡顿? - Android - 掘金 从这篇文章中你可以看到，Android 的...
passiontim 阅读 13,542 评论 2 赞 54
Android - 收藏集
用两张图告诉你，为
==========
Prompt: 2 tokens, 5.448 tokens-per-sec
Generation: 256 tokens, 42.701 tokens-per-sec
Peak memory: 49.838 GB
```