---
title: Mano-CUA-2.0-4B-MLX-8bit
canonical_url: "https://www.modelscope.cn/models/Mininglamp2718/Mano-CUA-2.0-4B-MLX-8bit"
md_url: "https://www.modelscope.cn/models/Mininglamp2718/Mano-CUA-2.0-4B-MLX-8bit.md"
repository: Mininglamp2718/Mano-CUA-2.0-4B-MLX-8bit
last_updated: 2026-06-24
license: apache-2.0
pipeline_tag: image-text-to-text
tasks:
- image-text-to-text
model_type:
- qwen3_vl
architectures:
- Qwen3VLForConditionalGeneration
base_model:
- Qwen/Qwen3-VL-4B
base_model_relation: quantized
parameters: 1.5B
tensor_type:
- U32
- F16
library_name:
- mlx
- safetensors
downloads: 205
stars: 2
tags:
- qwen3-vl
- vision-language
- gui-agent
- vla
- mano
- mlx
- 8bit
- quantized
---
# Mano-CUA-2.0-4B-MLX-8bit
> Mano-CUA-2.0-4B-MLX-8bit - Mininglamp2718 在 ModelScope 开源的模型。Mano-CUA-2.0-4B-MLX-8bit
Mininglamp2718/Mano-CUA-2.0-4B-MLX-8bit 是 ModelScope 魔搭社区上的 1.5B 参数image-text-to-text模型，采用 apache-2.0 许可，基于 Qwen/Qwen3-VL-4B 构建。
- **Repository**: Mininglamp2718/Mano-CUA-2.0-4B-MLX-8bit
- **License**: apache-2.0
- **Tasks**: image-text-to-text
- **Parameters**: 1.5B
- **Base model**: Qwen/Qwen3-VL-4B
- **Tags**: qwen3-vl, vision-language, gui-agent, vla, mano, mlx, 8bit, quantized
- **Downloads**: 205
- **Stars**: 2
- **Last updated**: 2026-06-24
Source: https://www.modelscope.cn/models/Mininglamp2718/Mano-CUA-2.0-4B-MLX-8bit
---
# Mano-CUA-2.0-4B-MLX-8bit
**Mano-CUA** is the Computer Use Agent model under the [Mano](https://github.com/Mininglamp-AI/Mano-P) open-source model series. It is a GUI-VLA (Visual Language Agent) model designed specifically for edge devices, capable of autonomously completing complex desktop GUI operations through visual understanding.
This is the **MLX 8-bit quantized** version, optimized for Apple Silicon (Mac mini / MacBook). For the full-precision fp16 version, see [Mano-CUA-2.0-4B](https://www.modelscope.cn/models/Mininglamp2718/Mano-CUA-2.0-4B).
## Main Capabilities
- **Complex GUI Automation**: Autonomously complete complex interface operations containing hundreds of interactive elements
- **Cross-System Data Integration**: Extract and integrate multi-source data through pure visual interaction without API interfaces
- **Long-Task Planning Execution**: Support enterprise-level business process automation of dozens to hundreds of steps
- **Intelligent Report Generation**: Automatically generate structured documents such as data analysis reports and work summaries
## Technical Background
Mano-CUA builds upon the complete technical framework of the Mano project (see [Mano Technical Report](https://arxiv.org/abs/2509.17336)), employing the Mano-Action bidirectional self-reinforcement learning method, three-stage progressive training (SFT → Offline Reinforcement Learning → Online Reinforcement Learning), "think-act-verify" loop reasoning mechanism, and a closed-loop data circulation system to achieve high-precision GUI understanding and operation capabilities. The edge version is optimized through mixed-precision quantization, visual token pruning, and edge inference adaptation, enabling large-scale parameter models to run efficiently on edge devices like Mac mini/MacBook/computing sticks.
## Quick Start
### Requirements
- macOS with Apple Silicon (M1+)
- Python >= 3.12
### Installation
**With Cider (recommended, includes W8A8 acceleration on M5+):**
```bash
pip install mlx-vlm
pip install git+https://github.com/Mininglamp-AI/cider.git
```
**Without Cider:**
```bash
pip install mlx-vlm
```
### Single-Step Demo
```python
import mlx_vlm as pm
from vlm_service import custom_generate
from PIL import Image
# 1. Load model
model, processor = pm.load("Mininglamp2718/Mano-CUA-2.0-4B-MLX-8bit")
# 2. Load a screenshot
img = Image.open("screenshot.png")
ratio = 1280 / img.width
img = img.resize((1280, int(img.height * ratio)), Image.LANCZOS)
# 3. Build prompt
task = "Click the search bar and type hello"
prompt_text = f"""You are a GUI agent. You are given a task and your action history, with screenshots. You need to perform the next action to complete the task.
## Output Format
action
## Action Space
open_app(app_name='') # Open an application by name.
open_url(url='') # Open a URL in the browser.
click(start_box='<|box_start|>(x1,y1)<|box_end|>')
type(content='') # type the content.
hotkey(key='') # Trigger a keyboard shortcut.
scroll(start_box='<|box_start|>(x1,y1)<|box_end|>', direction='down or up or right or left', amount='scroll_amount')
drag(start_box='<|box_start|>(x1,y1)<|box_end|>', end_box='<|box_start|>(x3,y3)<|box_end|>')
wait(duration='') # Sleep for specified duration (in seconds).
finish() # The task is completed.
stop(reason='') # If the item can not found in the image, give the reason
## User Instruction
{task}"""
messages = [
{"role": "system", "content": "You are a helpful assistant."},
{"role": "user", "content": prompt_text},
]
prompt = processor.tokenizer.apply_chat_template(
messages, tokenize=False, add_generation_prompt=True
)
prompt = prompt.replace("", "<|vision_start|><|image_pad|><|vision_end|>")
# 4. Run inference
result = custom_generate(
model, processor, prompt,
[img],
max_tokens=512,
temperature=0.0,
prefill_step_size=2048,
)
print(f"Tokens: {result.generation_tokens}, Speed: {result.generation_tps:.1f} tok/s")
print(result.text)
```
### Output Format
The model outputs structured XML:
```xml
The search bar is at the top of the page...
Click the search bar to focus it
click(start_box='<|box_start|>(500,38)<|box_end|>')
```
Coordinates are normalized to `[0, 1000]` range. To convert to pixel coordinates:
```python
pixel_x = int(x / 1000 * screen_width)
pixel_y = int(y / 1000 * screen_height)
```
### W8A8 Acceleration (M5+ only)
On Apple M5 or later, enable INT8 acceleration for ~15-19% faster prefill:
```python
from cider import convert_model, is_available
if is_available():
convert_model(model.language_model)
```
## Full Action Space
| Action | Syntax | Description |
| ------------ | ------------------------------------------------------------ | -------------------------- |
| open_app | `open_app(app_name='')` | Open an application |
| open_url | `open_url(url='')` | Open a URL |
| click | `click(start_box='<\|box_start\|>(x,y)<\|box_end\|>')` | Left click |
| doubleclick | `doubleclick(start_box='<\|box_start\|>(x,y)<\|box_end\|>')` | Double click |
| triple_click | `triple_click(start_box='<\|box_start\|>(x,y)<\|box_end\|>')` | Triple click (select line) |
| right_single | `right_single(start_box='<\|box_start\|>(x,y)<\|box_end\|>')` | Right click |
| hover | `hover(start_box='<\|box_start\|>(x,y)<\|box_end\|>')` | Mouse hover |
| type | `type(content='text')` | Type text |
| hotkey | `hotkey(key='cmd+c')` | Keyboard shortcut |
| hotkey_click | `hotkey_click(start_box='<\|box_start\|>(x,y)<\|box_end\|>', key='shift')` | Modifier + click |
| scroll | `scroll(start_box='<\|box_start\|>(x,y)<\|box_end\|>', direction='down', amount='3')` | Scroll |
| drag | `drag(start_box='<\|box_start\|>(x1,y1)<\|box_end\|>', end_box='<\|box_start\|>(x2,y2)<\|box_end\|>')` | Drag and drop |
| wait | `wait(duration='2')` | Wait (seconds) |
| finish | `finish()` | Task completed |
| stop | `stop(reason='...')` | Task infeasible |
| call_user | `call_user()` | Request human help |
## Other Versions
| Version | Repo | Description |
|---------|------|-------------|
| fp16 | [Mano-CUA-2.0-4B](https://www.modelscope.cn/models/Mininglamp2718/Mano-CUA-2.0-4B) | Full precision, for archival / re-quantization / GPU inference |
| MLX-8bit (this) | [Mano-CUA-2.0-4B-MLX-8bit](https://www.modelscope.cn/models/Mininglamp2718/Mano-CUA-2.0-4B-MLX-8bit) | MLX 8-bit quantized, recommended for Apple Silicon local inference |
## Contact
- Website: [https://github.com/Mininglamp-AI/Mano-P](https://github.com/Mininglamp-AI/Mano-P)
- Email: model@mininglamp.com