Instructions to use jackxinning/Leanly_AI with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use jackxinning/Leanly_AI with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="jackxinning/Leanly_AI",
	filename="Leanly_AI_14B_CHINESE_NONTHINK/Leanly_AI_14B_CHINESE_NONTHINK.gguf",
)

llm.create_chat_completion(
	messages = "{\n    \"question\": \"What is my name?\",\n    \"context\": \"My name is Clara and I live in Berkeley.\"\n}"
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use jackxinning/Leanly_AI with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf jackxinning/Leanly_AI
# Run inference directly in the terminal:
llama-cli -hf jackxinning/Leanly_AI

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf jackxinning/Leanly_AI
# Run inference directly in the terminal:
llama-cli -hf jackxinning/Leanly_AI

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 jackxinning/Leanly_AI
# Run inference directly in the terminal:
./llama-cli -hf jackxinning/Leanly_AI

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 jackxinning/Leanly_AI
# Run inference directly in the terminal:
./build/bin/llama-cli -hf jackxinning/Leanly_AI

Use Docker

docker model run hf.co/jackxinning/Leanly_AI

LM Studio
Jan
Ollama
How to use jackxinning/Leanly_AI with Ollama:
```
ollama run hf.co/jackxinning/Leanly_AI
```

Unsloth Studio new

How to use jackxinning/Leanly_AI 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 jackxinning/Leanly_AI 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 jackxinning/Leanly_AI to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for jackxinning/Leanly_AI to start chatting

Pi new

How to use jackxinning/Leanly_AI with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf jackxinning/Leanly_AI

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": "Leanly_AI"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use jackxinning/Leanly_AI with Docker Model Runner:
```
docker model run hf.co/jackxinning/Leanly_AI
```

Lemonade

How to use jackxinning/Leanly_AI with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull jackxinning/Leanly_AI

Run and chat with the model

lemonade run user.Leanly_AI-{{QUANT_TAG}}

List all available models

lemonade list

Leanly_AI Introduction

Leanly_AI is a large language model jointly developed by the Department of Endocrinology and Metabolism and the Department of General Practice at Provincial Hospital Affiliated to Fuzhou University.

Core Developers

Xin Ning
Wen Junping

Model Focus & Key Strengths

Leanly_AI is specifically designed for clinical weight management settings. It aims to assist physicians in providing standardized, supportive, and clinically oriented psychological care for individuals with obesity who experience emotional challenges during the weight management process.

A key feature of Leanly_AI is its deep integration with large-scale population health evidence from the UK Biobank. This integration enables the model to better understand the psychological characteristics, emotional vulnerability, and clinical support needs of individuals with obesity, while maintaining a clinically interpretable and evidence-informed response style.

Unlike general conversational models, Leanly_AI focuses on the psychological needs commonly encountered in obesity and weight management, including anxiety, low mood, guilt, self-blame, emotional distress, frustration, weight-related stigma, and reduced treatment motivation.

For more information, please visit our official website: http://www.leanly-ai.top/

From General Psychological Comfort to Evidence-Informed Clinical Psychological Support

Traditional psychological comfort models often provide broad encouragement, such as “Don’t be sad,” “Keep going,” or “You can do it.” However, individuals with obesity may face more complex emotional burdens during weight management, including repeated treatment failure, body image distress, social pressure, stigma, and reduced self-efficacy.

Leanly_AI is designed to move beyond simple emotional comfort. By incorporating insights from UK Biobank-based population health data and clinical weight management scenarios, the model provides responses that are more structured, context-aware, clinically cautious, and aligned with real-world obesity care.

For example, when a user expresses persistent sadness, obvious anxiety, feelings of worthlessness, emotional loss of control, or reduced motivation for treatment, the model does not merely provide reassurance. Instead, it can help identify potential psychological distress, encourage standardized psychological screening when appropriate, and remind users to seek support from physicians or mental health professionals when necessary.

Clinical Logic and Medical Interpretability

The design of Leanly_AI emphasizes clinical logic and medical interpretability. Its response strategy is not limited to general language generation. Instead, it is informed by large-scale population-level health evidence, clinical medical knowledge, and the psychological characteristics commonly observed in individuals undergoing obesity treatment.

Within this framework, the model pays special attention to:

Emotional problems that may interfere with weight management
Psychological distress related to weight stigma and body image concerns
Situations that may require professional psychological evaluation
Communication strategies that reduce shame, guilt, and self-blame
Supportive guidance that improves treatment adherence and long-term engagement

Therefore, Leanly_AI is not only an emotional companionship tool, but also a potential auxiliary tool for early psychological risk awareness, stigma-reducing communication, treatment adherence support, and clinically oriented weight management care.

Scientific Innovation

From a research and application perspective, Leanly_AI explores a new direction for integrating large-scale biomedical data resources, medical knowledge, clinical experience, and AI-based psychological support in the field of obesity care.

Most existing AI psychological support models are built for general emotional conversation. In contrast, Leanly_AI is tailored to a specific clinical population and a specific healthcare scenario: individuals with obesity during weight management. Its development is supported by a deep fusion of UK Biobank-derived population health evidence and clinically oriented psychological support strategies.

This gives the model stronger:

Disease-population specificity
Clinical scenario adaptability
Evidence-informed response logic
Medical communication awareness
Stigma-reducing response style
Potential value in physician-assisted care

The goal of Leanly_AI is not to replace physicians or mental health professionals, but to provide a supportive AI tool that helps improve psychological care during obesity management.

Key Advantages

✅ Deeply integrates UK Biobank-based population health evidence
✅ Provides standardized psychological support for individuals with obesity
✅ Helps reduce psychological harm caused by weight-related stigma
✅ Supports early awareness of possible emotional distress
✅ Encourages appropriate psychological screening and professional help-seeking
✅ Improves patient treatment adherence and long-term engagement
✅ Enhances the overall quality of weight management care
✅ Assists physicians in providing more structured and empathetic communication
✅ Focuses on clinically relevant emotional problems during obesity treatment

Core Feature Summary

The core feature of Leanly_AI can be summarized as follows:

Leanly_AI is a clinically oriented psychological support large language model for individuals with obesity. It deeply integrates UK Biobank-based population health evidence, medical knowledge, weight management scenarios, and stigma-reducing communication strategies to provide structured, empathetic, and physician-assisted psychological care during the weight management process.

Leanly_AI — Providing professional psychological support to make weight management more comprehensive, personalized, and sustainable.

Downloads last month: 10,330

GGUF

Model size

15B params

Architecture

qwen3

Hardware compatibility

We're not able to determine the quantization variants.

View all variants