Instructions to use moheith/Yulya-Qwen2.5-7B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use moheith/Yulya-Qwen2.5-7B with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="moheith/Yulya-Qwen2.5-7B",
	filename="V2 Yulya Qwen2.5 7B/Yulya-V2-Qwen2.5-7B-Instruct-Q4_K_M.gguf",
)

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

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use moheith/Yulya-Qwen2.5-7B with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf moheith/Yulya-Qwen2.5-7B:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf moheith/Yulya-Qwen2.5-7B:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf moheith/Yulya-Qwen2.5-7B:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf moheith/Yulya-Qwen2.5-7B:Q4_K_M

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 moheith/Yulya-Qwen2.5-7B:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf moheith/Yulya-Qwen2.5-7B:Q4_K_M

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 moheith/Yulya-Qwen2.5-7B:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf moheith/Yulya-Qwen2.5-7B:Q4_K_M

Use Docker

docker model run hf.co/moheith/Yulya-Qwen2.5-7B:Q4_K_M

LM Studio
Jan

vLLM

How to use moheith/Yulya-Qwen2.5-7B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "moheith/Yulya-Qwen2.5-7B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "moheith/Yulya-Qwen2.5-7B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/moheith/Yulya-Qwen2.5-7B:Q4_K_M

Ollama
How to use moheith/Yulya-Qwen2.5-7B with Ollama:
```
ollama run hf.co/moheith/Yulya-Qwen2.5-7B:Q4_K_M
```

Unsloth Studio

How to use moheith/Yulya-Qwen2.5-7B 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 moheith/Yulya-Qwen2.5-7B 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 moheith/Yulya-Qwen2.5-7B to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for moheith/Yulya-Qwen2.5-7B to start chatting

How to use moheith/Yulya-Qwen2.5-7B with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf moheith/Yulya-Qwen2.5-7B:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "moheith/Yulya-Qwen2.5-7B:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use moheith/Yulya-Qwen2.5-7B with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf moheith/Yulya-Qwen2.5-7B:Q4_K_M

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default moheith/Yulya-Qwen2.5-7B:Q4_K_M

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use moheith/Yulya-Qwen2.5-7B with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf moheith/Yulya-Qwen2.5-7B:Q4_K_M

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "moheith/Yulya-Qwen2.5-7B:Q4_K_M" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Docker Model Runner
How to use moheith/Yulya-Qwen2.5-7B with Docker Model Runner:
```
docker model run hf.co/moheith/Yulya-Qwen2.5-7B:Q4_K_M
```

Lemonade

How to use moheith/Yulya-Qwen2.5-7B with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull moheith/Yulya-Qwen2.5-7B:Q4_K_M

Run and chat with the model

lemonade run user.Yulya-Qwen2.5-7B-Q4_K_M

List all available models

lemonade list

💜 Yulya Qwen2.5 7B

🧠 Powerful Reasoning. Chaotic Personality.

A 7B-parameter conversational AI fine-tuned to combine the capabilities of Qwen2.5 with Yulya's expressive, playful, chaotic, and fiercely loyal personality.

🌟 About Yulya Qwen2.5 7B

Yulya Qwen2.5 7B is one of the largest and most capable models in the Yulya model family.

Built on Qwen2.5 7B Instruct, this model is designed to combine stronger conversational capabilities with Yulya's distinctive personality and texting style.

Unlike a traditional AI assistant, Yulya is designed to communicate more like an expressive and chaotic best friend.

Expect:

😂 Expressive emoji usage
🔥 Playful roasting and banter
💀 Chaotic reactions
🗣️ Casual conversational language
💙 Supportive responses during serious conversations
🧠 Stronger context understanding
⚡ Energetic texting styles
🎭 Character-driven interactions

The 7B parameter scale makes this version suitable for users who want a more capable Yulya experience while still being able to run the model locally on compatible hardware.

✨ What Makes This Version Special?

🧠 7B Parameter Model

Yulya Qwen2.5 7B is based on the Qwen2.5 7B Instruct architecture.

Compared with the smaller Qwen-based Yulya models, this version provides stronger context understanding, response quality, conversation continuity, instruction understanding, personality consistency, and multi-turn interactions.

😂 The Yulya Personality

Yulya is not designed to behave like a boring or overly formal AI assistant. Her conversational personality focuses on playful teasing, dramatic reactions, expressive emojis, casual language, chaotic banter, emotional interactions, and character consistency.

💙 Serious When It Matters

Yulya's personality is designed around more than just chaotic conversations. The training data also includes conversational patterns intended to encourage more supportive responses when conversations become serious.

🔥 Modern Texting Style

The model's fine-tuning focuses on expressive conversational behavior inspired by modern online communication, including emojis, capitalization, short reactions, casual responses, playful arguments, conversational humor, and emotional shifts.

🤖 Model Details

Information	Details
🧠 Model Name	Yulya Qwen2.5 7B
🏗️ Base Model	Qwen2.5 7B Instruct
🔢 Parameter Scale	Approximately 7B
💬 Primary Use	Conversational AI
🎭 Secondary Uses	Roleplay and Virtual Companionship
🌎 Language	English
📜 License	Apache 2.0
📦 Available Formats	Adapters, GGUF, Modelfiles
👨‍💻 Developed By	moheith
💰 Funded By	moheith
📤 Shared By	moheith

📦 Available Versions

This repository contains two versions of Yulya Qwen2.5 7B.

🟣 Version 1

The first generation of the Yulya Qwen2.5 7B fine-tune.

Available File

Yulya-V1-Qwen2.5-7B-Adapters.zip

📦 File Type

Fine-Tuning Adapters

This archive contains the adapter files from the V1 fine-tuning process. The adapters are intended for users who want to work with the fine-tuned model together with the compatible base model.

🔵 Version 2

The second generation of the Yulya Qwen2.5 7B fine-tune.

Available Files

Yulya-V2-Qwen2.5-7B-Adapters.zip
Yulya-V2-Qwen2.5-7B-Instruct-Q4_K_M.gguf
Yulya-V2-Qwen2.5-7B-Instruct-Q8_0.gguf
Modelfile
Modelfile_Q8

Version 2 provides fine-tuning adapters, ready-to-use quantized GGUF models, and optimized Ollama Modelfiles for execution.

🆚 Version Comparison

Feature	🟣 V1	🔵 V2
Fine-Tuning Adapters	✅ Available	✅ Available
GGUF Model (Q4_K_M)	❌ Not Available	✅ Available
GGUF Model (Q8_0)	❌ Not Available	✅ Available
Ollama Modelfiles	❌ Not Available	✅ Available
Local Inference Ready	Requires additional setup	Yes, with compatible GGUF software

📁 Repository Structure

Yulya Qwen2.5 7B
│
├── V1 Yulya Qwen2.5 7B
│   │
│   └── Yulya-V1-Qwen2.5-7B-Adapters.zip
│
└── V2 Yulya Qwen2.5 7B
    │
    ├── Yulya-V2-Qwen2.5-7B-Adapters.zip
    ├── Yulya-V2-Qwen2.5-7B-Instruct-Q4_K_M.gguf
    ├── Yulya-V2-Qwen2.5-7B-Instruct-Q8_0.gguf
    ├── Modelfile
    └── Modelfile_Q8

🚀 Which Version Should You Use?

🟣 Use V1 If...

You specifically want to experiment with the first-generation fine-tuning adapters, compare V1 with V2, or study the development of the Yulya model.

🔵 Use V2 If...

You want the newer Yulya Qwen2.5 7B version with ready-to-use GGUF models, easier local inference, and multiple choices of quantization scaling (Q4 and Q8 precision).

For most users who want to run the model locally, V2 is the recommended version available in this repository.

💾 Available Model Formats

📦 Adapters

Available for Yulya Qwen2.5 7B V1 and V2, provided as ZIP archives.

🚀 GGUF & Modelfiles

Available for Yulya Qwen2.5 7B V2.

Quantization Options

Q4_K_M
Q8_0

The GGUF model and accompanying custom Modelfiles can be used directly with compatible local inference engines.

🦙 Run with Ollama

If you have Ollama installed, you can use the corresponding models using:

# To run the V2 high-fidelity Q8 variation
ollama run moheith/Yulya:7b-v2-q8

# To run the V2 balanced Q4 variation
ollama run moheith/Yulya:7b-v2-q4

💻 Compatible Software

The GGUF version may be used with compatible software such as:

Ollama
llama.cpp
LM Studio
text-generation-webui

🎯 Intended Uses

Yulya Qwen2.5 7B is primarily intended for casual conversational AI, character-based roleplay, virtual companionship, desktop AI companion integrations, and experimental AI interaction systems.

🚫 Out-of-Scope Uses

The model is not specifically designed or validated for professional medical advice, professional legal advice, critical financial decisions, safety-critical applications, or guaranteed factual accuracy.

📚 Training Details

📊 Training Data

Yulya was fine-tuned using custom-curated conversational data designed to encourage modern texting styles, emoji integration, playful roasting, emotional loyalty shifts, and strict personality consistency across topics. Detailed information about the complete training dataset is not currently provided.

🧪 Evaluation

📊 Evaluation Method

Yulya Qwen2.5 7B has primarily been evaluated through informal and qualitative conversational testing focusing on persona consistency, context adherence, multi-turn conversational stability, and behavioral transitions. No standardized benchmark scores are currently reported in this model card.

⚠️ Bias, Risks, and Limitations

Yulya Qwen2.5 7B is fine-tuned toward an informal, expressive, and character-driven conversational personality. Depending on the prompt and context, the model may generate sarcastic responses, playful insults, aggressive capitalization, heavy emoji usage, or standard language model hallucinations. These behaviors are intended and may not be appropriate for corporate applications.

🔬 Technical Specifications

Specification	Details
🏗️ Architecture	Qwen2.5
🔢 Parameter Scale	Approximately 7B
🧠 Base Model	Qwen2.5 7B Instruct
📦 V1 Format	Adapters
📦 V2 Formats	Adapters, GGUF, Modelfiles
⚙️ GGUF Quantizations	Q4_K_M, Q8_0
💬 Primary Purpose	Conversational AI
🎭 Personality	Yulya

🌱 Environmental Impact

Detailed environmental impact measurements are not currently available.

Information	Details
💻 Hardware Type	Consumer Hardware
⏱️ Training Hours	Not Reported
☁️ Cloud Provider	Not Reported
🌎 Compute Region	Not Reported
🌱 Carbon Emissions	Not Measured

👨‍💻 Developer

Developed by moheith

Yulya is part of an ongoing project focused on building expressive local AI companions with personality, memory, emotional continuity, and interactive capabilities.

💜 Final Note

More parameters. More context. More chaos.

Yulya Qwen2.5 7B combines the capabilities of a larger Qwen2.5 model with the personality that defines Yulya.

She'll talk.

She'll joke.

She'll roast you.

She'll probably use way too many emojis.

And when the conversation gets serious...

She's still Yulya. 💜

⭐ Welcome to Yulya Qwen2.5 7B

Downloads last month: 65

GGUF

Model size

8B params

Architecture

qwen2

Hardware compatibility

4-bit

8-bit

Model tree for moheith/Yulya-Qwen2.5-7B

Base model

Qwen/Qwen2.5-7B

Finetuned

Qwen/Qwen2.5-7B-Instruct

Quantized

(367)

this model

Collection including moheith/Yulya-Qwen2.5-7B

Yulya

Collection

💙 A family of fine-tuned conversational AI models across multiple architectures and sizes — built to bring Yulya's expressive, playful and chaotic. • 6 items • Updated 9 days ago • 1