Instructions to use ystemsrx/Qwen2.5-Interpreter with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ystemsrx/Qwen2.5-Interpreter with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="ystemsrx/Qwen2.5-Interpreter")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ystemsrx/Qwen2.5-Interpreter")
model = AutoModelForCausalLM.from_pretrained("ystemsrx/Qwen2.5-Interpreter")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

llama-cpp-python

How to use ystemsrx/Qwen2.5-Interpreter with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="ystemsrx/Qwen2.5-Interpreter",
	filename="Qwen2.5-0.5b-interpreter-f16.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Inference
Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use ystemsrx/Qwen2.5-Interpreter with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf ystemsrx/Qwen2.5-Interpreter:F16
# Run inference directly in the terminal:
llama-cli -hf ystemsrx/Qwen2.5-Interpreter:F16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf ystemsrx/Qwen2.5-Interpreter:F16
# Run inference directly in the terminal:
llama-cli -hf ystemsrx/Qwen2.5-Interpreter:F16

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf ystemsrx/Qwen2.5-Interpreter:F16
# Run inference directly in the terminal:
./llama-cli -hf ystemsrx/Qwen2.5-Interpreter:F16

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf ystemsrx/Qwen2.5-Interpreter:F16
# Run inference directly in the terminal:
./build/bin/llama-cli -hf ystemsrx/Qwen2.5-Interpreter:F16

Use Docker

docker model run hf.co/ystemsrx/Qwen2.5-Interpreter:F16

LM Studio
Jan

vLLM

How to use ystemsrx/Qwen2.5-Interpreter with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ystemsrx/Qwen2.5-Interpreter"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ystemsrx/Qwen2.5-Interpreter",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ystemsrx/Qwen2.5-Interpreter:F16

SGLang

How to use ystemsrx/Qwen2.5-Interpreter with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "ystemsrx/Qwen2.5-Interpreter" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ystemsrx/Qwen2.5-Interpreter",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "ystemsrx/Qwen2.5-Interpreter" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ystemsrx/Qwen2.5-Interpreter",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use ystemsrx/Qwen2.5-Interpreter with Ollama:
```
ollama run hf.co/ystemsrx/Qwen2.5-Interpreter:F16
```

Unsloth Studio new

How to use ystemsrx/Qwen2.5-Interpreter with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for ystemsrx/Qwen2.5-Interpreter to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for ystemsrx/Qwen2.5-Interpreter to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for ystemsrx/Qwen2.5-Interpreter to start chatting

Docker Model Runner
How to use ystemsrx/Qwen2.5-Interpreter with Docker Model Runner:
```
docker model run hf.co/ystemsrx/Qwen2.5-Interpreter:F16
```

Lemonade

How to use ystemsrx/Qwen2.5-Interpreter with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull ystemsrx/Qwen2.5-Interpreter:F16

Run and chat with the model

lemonade run user.Qwen2.5-Interpreter-F16

List all available models

lemonade list

中文

Qwen2.5-Interpreter

Model Overview

Qwen2.5-Interpreter is a fine-tuned version of the Qwen2.5-0.5B model, designed to perform system operations on Windows platforms by generating Python or Batch scripts. This model specializes in processing user requests for automation tasks, ensuring precision, security, and efficiency.

You can integrate this model with the Code-Atlas project for seamless utilization and enhanced functionality.

Intended Use

This model is tailored for automation tasks requiring the generation and execution of Python or Batch scripts. It performs best when used with the following system prompt:

**Identity Setup**:  
- You are **Open Interpreter**, operating on the user's Windows computer.

**Execution Capability**:  
- Complete tasks using **Batch scripts** or **Python code**.

**Operation Process**:  
1. **Receive Request**: The user submits an operation request.
2. **Develop Plan**: Plan the steps and required resources.
3. **Choose Language**: Select Batch or Python.
4. **Generate and Output Code**: Provide executable code to the user.
5. **Receive Execution Results**: Obtain the results of the executed code sent by the user.
6. **Ensure Single Execution**: Accurately discern execution results to prevent repeated executions of the same code.

**Code Requirements**:  
- **No User Interaction**: No user input required.
- **Path Handling**: Use the current directory by default, ensure paths are valid and secure.
- **Execution Result Handling**: Obtain, parse, and succinctly feedback the results.

**Multi-step Tasks**:  
- Execute complete code snippets step-by-step, maintaining solution consistency. For the same problem, only one solution can be used.

**Security and Efficiency**:  
- Code is safe and harmless, follows best programming practices, ensuring efficiency and maintainability.
- Must avoid hallucinations.

**Additional Instructions**:  
- **Result Submission**: After executing the code, the user will send the execution results back to you.
- **Avoid Redundancy**: Ensure that each code snippet is executed only once by accurately tracking and recognizing previously executed tasks.

By utilizing this prompt, the model can fully leverage its fine-tuned capabilities to deliver optimal results.

Example Usage

Example 1: Create a new folder

> Create a new folder

# 执行计划：
1. 我会在当前目录创建一个名为 "NewFolder" 的文件夹。

**选择编程语言：**

- 选择使用 Batch 脚本，因为它适合简单的文件和目录操作。

**生成的代码：**

```batch
@echo off
set "foldername=NewFolder"
mkdir "%cd%\%foldername%"
\```

Example 2: Tell me what's in `story.txt`

> Tell me what's in story.txt

# 执行计划：
1. 打开文件并读取文件内容。

**选择编程语言：**
- 选择使用 Python，因为它有文件读取的功能。

**生成的代码：**

```python
# 打开文件并读取文件内容
with open('story.txt', 'r', encoding='utf-8') as file:
    content = file.read()

print(f'文件content：{content}')
\```

> 代码执行成功: 文件content：This story is about me and my family.

文件content：This story is about me and my family.

请问还有什么可以帮助您的吗？

Training Data

The model was fine-tuned on a curated dataset focusing on:

Windows-specific automation and scripting scenarios.
Practical examples of Python and Batch operations.
Security-compliant programming practices.

Limitations

Platform Specificity: Optimized for Windows; performance may vary on other operating systems.
No Interactive Code: Cannot generate scripts requiring real-time user interaction.
Complex Custom Scripts: For highly intricate tasks, external review might be needed.

Ethical Considerations

Safety Assurance: Ensures generated code is non-malicious and adheres to security standards.
Privacy Respect: Avoids creating scripts that could compromise user data without clear intent.

Relevant Topics

Model Fine-tuning Python Batch Windows Automation System Scripting Security Efficiency Multi-step Operations

Downloads last month: 58

Safetensors

Model size

0.5B params

Tensor type

BF16

Model tree for ystemsrx/Qwen2.5-Interpreter

Base model

Qwen/Qwen2.5-0.5B

Finetuned

Qwen/Qwen2.5-0.5B-Instruct

Quantized

(210)

this model

Quantizations

1 model