Instructions to use shiqianyan/SafetyRAISE-MainAgent-Exp with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use shiqianyan/SafetyRAISE-MainAgent-Exp 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 shiqianyan/SafetyRAISE-MainAgent-Exp:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf shiqianyan/SafetyRAISE-MainAgent-Exp:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf shiqianyan/SafetyRAISE-MainAgent-Exp:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf shiqianyan/SafetyRAISE-MainAgent-Exp:Q4_K_M

Build from source code

git clone https://github.com/ggml-org/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-app
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama serve -hf shiqianyan/SafetyRAISE-MainAgent-Exp:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama cli -hf shiqianyan/SafetyRAISE-MainAgent-Exp:Q4_K_M

Use Docker

docker model run hf.co/shiqianyan/SafetyRAISE-MainAgent-Exp:Q4_K_M

Llama new
LM Studio
Jan
Ollama
How to use shiqianyan/SafetyRAISE-MainAgent-Exp with Ollama:
```
ollama run hf.co/shiqianyan/SafetyRAISE-MainAgent-Exp:Q4_K_M
```
Unsloth Desktop

How to use shiqianyan/SafetyRAISE-MainAgent-Exp with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf shiqianyan/SafetyRAISE-MainAgent-Exp:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "shiqianyan/SafetyRAISE-MainAgent-Exp:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use shiqianyan/SafetyRAISE-MainAgent-Exp with Docker Model Runner:
```
docker model run hf.co/shiqianyan/SafetyRAISE-MainAgent-Exp:Q4_K_M
```

Lemonade

How to use shiqianyan/SafetyRAISE-MainAgent-Exp with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull shiqianyan/SafetyRAISE-MainAgent-Exp:Q4_K_M

Run and chat with the model

lemonade run user.SafetyRAISE-MainAgent-Exp-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use shiqianyan/SafetyRAISE-MainAgent-Exp with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf shiqianyan/SafetyRAISE-MainAgent-Exp: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 shiqianyan/SafetyRAISE-MainAgent-Exp:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use shiqianyan/SafetyRAISE-MainAgent-Exp with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf shiqianyan/SafetyRAISE-MainAgent-Exp: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 "shiqianyan/SafetyRAISE-MainAgent-Exp: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"

SafetyRAISE-MainAgent-Exp

这是一个实验仓库，存放 SafetyRAISE 主智能体实验产出的量化模型（GGUF）。按实验分子目录，每个子目录里有模型文件和配套的运行文件（聊天模板、思考预算提示语、启动示例），细节见各子目录的 README。

目录	内容
`report-exp-v1/`	报告实验第一版：Qwen3.8-27B 经 LoRA 微调与偏好优化后合并量化（Q4_K_M），附重训的 MTP 头
`report-exp-v2/`	报告实验第二版：在第一版权重上做 MLP 方向编辑后重新量化（Q4_K_M），附在编辑后模型上重训的 MTP 头

暂不开放原始权重：LoRA 适配器、单独的 MTP 头权重、编辑参数和全精度权重都不在这里。

训练与评测代码、数据说明和技术报告见 GitHub 仓库：https://github.com/QianYan-Art/SafetyRAISE-Lora

按 Apache License 2.0 发布（见 LICENSE、NOTICE），与基座 Qwen3.8-27B 相同。

Downloads last month: 70

GGUF

Model size

27B params

Architecture

qwen35

Hardware compatibility

4-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for shiqianyan/SafetyRAISE-MainAgent-Exp

Base model

Qwen/Qwen3.8-27B

Quantized

(1,681)

this model