Instructions to use gatilin/Marvis-MLocalModel3.6.2-35B-Q4_K_M with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use gatilin/Marvis-MLocalModel3.6.2-35B-Q4_K_M 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 gatilin/Marvis-MLocalModel3.6.2-35B-Q4_K_M:BF16
# Run inference directly in the terminal:
llama cli -hf gatilin/Marvis-MLocalModel3.6.2-35B-Q4_K_M:BF16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf gatilin/Marvis-MLocalModel3.6.2-35B-Q4_K_M:BF16
# Run inference directly in the terminal:
llama cli -hf gatilin/Marvis-MLocalModel3.6.2-35B-Q4_K_M:BF16

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 gatilin/Marvis-MLocalModel3.6.2-35B-Q4_K_M:BF16
# Run inference directly in the terminal:
./llama-cli -hf gatilin/Marvis-MLocalModel3.6.2-35B-Q4_K_M:BF16

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 gatilin/Marvis-MLocalModel3.6.2-35B-Q4_K_M:BF16
# Run inference directly in the terminal:
./build/bin/llama-cli -hf gatilin/Marvis-MLocalModel3.6.2-35B-Q4_K_M:BF16

Use Docker

docker model run hf.co/gatilin/Marvis-MLocalModel3.6.2-35B-Q4_K_M:BF16

LM Studio
Jan
Ollama
How to use gatilin/Marvis-MLocalModel3.6.2-35B-Q4_K_M with Ollama:
```
ollama run hf.co/gatilin/Marvis-MLocalModel3.6.2-35B-Q4_K_M:BF16
```

Unsloth Studio

How to use gatilin/Marvis-MLocalModel3.6.2-35B-Q4_K_M 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 gatilin/Marvis-MLocalModel3.6.2-35B-Q4_K_M 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 gatilin/Marvis-MLocalModel3.6.2-35B-Q4_K_M to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for gatilin/Marvis-MLocalModel3.6.2-35B-Q4_K_M to start chatting

How to use gatilin/Marvis-MLocalModel3.6.2-35B-Q4_K_M with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf gatilin/Marvis-MLocalModel3.6.2-35B-Q4_K_M:BF16

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": "gatilin/Marvis-MLocalModel3.6.2-35B-Q4_K_M:BF16"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use gatilin/Marvis-MLocalModel3.6.2-35B-Q4_K_M with Docker Model Runner:
```
docker model run hf.co/gatilin/Marvis-MLocalModel3.6.2-35B-Q4_K_M:BF16
```

Lemonade

How to use gatilin/Marvis-MLocalModel3.6.2-35B-Q4_K_M with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull gatilin/Marvis-MLocalModel3.6.2-35B-Q4_K_M:BF16

Run and chat with the model

lemonade run user.Marvis-MLocalModel3.6.2-35B-Q4_K_M-BF16

List all available models

lemonade list

Hermes Agent

How to use gatilin/Marvis-MLocalModel3.6.2-35B-Q4_K_M with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf gatilin/Marvis-MLocalModel3.6.2-35B-Q4_K_M:BF16

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 gatilin/Marvis-MLocalModel3.6.2-35B-Q4_K_M:BF16

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use gatilin/Marvis-MLocalModel3.6.2-35B-Q4_K_M with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf gatilin/Marvis-MLocalModel3.6.2-35B-Q4_K_M:BF16

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 "gatilin/Marvis-MLocalModel3.6.2-35B-Q4_K_M:BF16" \
  --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"

YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

Marvis 本地模型权重导出包

导出时间：2026-07-21 来源：/Users/Library/Application Support/com.tencent.mac.marvis/llm-sdk/

内容说明

models/mLocalModel3.6.2.gguf — 主语言模型权重（35B Q4_K_M，约 15 GB）
models/vision/mmproj-BF16.gguf — 视觉多模态投影权重（约 861 MB）
engines/ — llama.cpp Metal 后端推理引擎及动态库
state/ — 运行状态 JSON（llm_state.json、download_state.json）
config/ — 设备匹配缓存（device_match_cache.json）

模型信息

引擎：llama.cpp-metal，版本 b9846
模型：MLocalModel3.6.2 35B Q4_K_M
量化：Q4_K_M
格式：GGUF

原生运行参数示例：

--host 127.0.0.1 -fa 1 -ctk q4_0 -ctv q4_0 --reasoning on \
--log-prefix --log-timestamps --log-colors off --fit on \
-np 3 -c 393216 --no-mmap --fit-target 3200 \
-m models/mLocalModel3.6.2.gguf -t 8 \
--mmproj models/vision/mmproj-BF16.gguf

注意

本包仅供个人备份/迁移使用。模型权重受腾讯/Marvis 相关服务条款约束。

Downloads last month: 5

GGUF

Model size

27B params

Architecture

qwen35moe

Hardware compatibility

We're not able to determine the quantization variants.

View all variants

Inference Providers NEW

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