Instructions to use giabaoboon/vmax-personal-core-models with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use giabaoboon/vmax-personal-core-models 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 giabaoboon/vmax-personal-core-models:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf giabaoboon/vmax-personal-core-models:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf giabaoboon/vmax-personal-core-models:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf giabaoboon/vmax-personal-core-models: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 giabaoboon/vmax-personal-core-models:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf giabaoboon/vmax-personal-core-models: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 giabaoboon/vmax-personal-core-models:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf giabaoboon/vmax-personal-core-models:Q4_K_M

Use Docker

docker model run hf.co/giabaoboon/vmax-personal-core-models:Q4_K_M

LM Studio
Jan
Ollama
How to use giabaoboon/vmax-personal-core-models with Ollama:
```
ollama run hf.co/giabaoboon/vmax-personal-core-models:Q4_K_M
```

Unsloth Studio

How to use giabaoboon/vmax-personal-core-models 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 giabaoboon/vmax-personal-core-models 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 giabaoboon/vmax-personal-core-models to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for giabaoboon/vmax-personal-core-models to start chatting

How to use giabaoboon/vmax-personal-core-models with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf giabaoboon/vmax-personal-core-models: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": "giabaoboon/vmax-personal-core-models:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use giabaoboon/vmax-personal-core-models with Docker Model Runner:
```
docker model run hf.co/giabaoboon/vmax-personal-core-models:Q4_K_M
```

Lemonade

How to use giabaoboon/vmax-personal-core-models with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull giabaoboon/vmax-personal-core-models:Q4_K_M

Run and chat with the model

lemonade run user.vmax-personal-core-models-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use giabaoboon/vmax-personal-core-models with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf giabaoboon/vmax-personal-core-models: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 giabaoboon/vmax-personal-core-models:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use giabaoboon/vmax-personal-core-models with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf giabaoboon/vmax-personal-core-models: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 "giabaoboon/vmax-personal-core-models: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"

VMAX Personal Core — models and Windows release

Source repository: Codeberg link will be published after owner authentication.

This repository distributes the large runtime assets used by VMAX 2.2.0. Source code is maintained separately on Codeberg so it can be forked and edited without pulling multi-gigabyte model weights.

Components

Component	Upstream/base	License/provenance	SHA-256
Personal Vietnamese ASR CT2	`vinai/PhoWhisper-base` family	PhoWhisper base: BSD-3-Clause; VMAX personal fine-tune published with owner authorization	`B1FDB6A9294142D5728A36ACDD563040ED5CB802E26721F360819918981E9F74` (`model.bin`)
VMAX Report Agent Q4_K_M	`unsloth/Qwen3-4B-Instruct-2507-unsloth-bnb-4bit`	Apache-2.0 base; project-generated/user-owned SFT data	`A5B7D6462EBC6573F808D36E095CADE36C3EF81E7D4D0E4AE1EC61802B19C345`
Vietnamese Piper voice	`rhasspy/piper-voices`, `vi_VN-vais1000-medium`	MIT repository; VAIS-1000 dataset CC BY 4.0	`EC7C89E2C85F4D1EDC24B6120C18AAF1BDA614F06B511567EB9C7C0DE15E2DAB`
Wake/stop classifiers	VMAX project assets	User-owned project assets	See `SHA256SUMS.txt`
VMAX source	VMAX contributors	MIT	See Codeberg source repository

The differently named UNVERIFIED report GGUF was not uploaded: it was an exact duplicate of the canonical GGUF (same SHA-256), not a second model.

Privacy and API keys

No API key, Windows credential, user database, conversation log, audio recording, or learning data is included. VMAX requires every end user to configure their own Gemini API key. The Windows app stores it using Windows DPAPI.

Install

Download and extract the Windows Core ZIP from releases/.
Download the folders under models/ into the matching VMAX/models/ paths.
Extract runtime/VMAX-2.2.0-TTS-Native-Windows-x64-20260813.zip into VMAX/.runtime/; the archive already contains the tts-native/ folder.
Follow docs/DISTRIBUTION_A_TO_Z_VI.md in the release.
Run VMAX.exe api-key, then VMAX.exe doctor and VMAX.exe voice.

Limitations

Windows x64 only.
Personal ASR can perform differently for voices outside its fine-tuning data.
Report Agent output must still pass VMAX's local schema validation and document postflight.
Users are responsible for the quota/cost and terms of their own API account.

See THIRD_PARTY_NOTICES_TTS.md and the source repository for complete notices.

Downloads last month: 7

GGUF

Model size

4B params

Architecture

qwen3

Hardware compatibility

4-bit