Instructions to use GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF",
	filename="Hy-MT2-30B-A3B-Q2_K.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 GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF: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 GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF: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 GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF:Q4_K_M

Use Docker

docker model run hf.co/GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF:Q4_K_M

Ollama
How to use GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF with Ollama:
```
ollama run hf.co/GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF:Q4_K_M
```

Unsloth Studio

How to use GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF 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 GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF 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 GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF to start chatting

How to use GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF: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": "GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF: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 GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF:Q4_K_M

Run Hermes

hermes

Docker Model Runner
How to use GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF with Docker Model Runner:
```
docker model run hf.co/GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF:Q4_K_M
```

Lemonade

How to use GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Hy-MT2-30B-A3B-4bit-GGUF-Q4_K_M

List all available models

lemonade list

Hy-MT2-30B-A3B 4bit GGUF

GGUF quantizations of tencent/Hy-MT2-30B-A3B, with Q4_K_M as the recommended 4bit deployment file.

Files

File	Size	Notes
`Hy-MT2-30B-A3B-Q2_K.gguf`	10.30 GiB	smallest test build
`Hy-MT2-30B-A3B-Q3_K_M.gguf`	13.45 GiB	lower VRAM option
`Hy-MT2-30B-A3B-Q4_K_M.gguf`	16.98 GiB	recommended default
`Hy-MT2-30B-A3B-BF16.gguf`	56.03 GiB	conversion source
`patches/llama-cpp-hyv3.patch`	17 KiB	required when upstream llama.cpp has no HYV3 support

Important

Current upstream llama.cpp may not support HYV3ForCausalLM / model_type=hy_v3 directly. This repository includes the required patch at:

patches/llama-cpp-hyv3.patch

Clone this model repository first, then prepare a patched llama.cpp checkout:

git lfs install
git clone https://huggingface.co/GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF
cd Hy-MT2-30B-A3B-4bit-GGUF
bash scripts/prepare_llama_cpp.sh ./llama.cpp

For CUDA 13.1 builds:

CUDA_HOME=/usr/local/cuda-13.1 bash scripts/prepare_llama_cpp.sh ./llama.cpp

Run Q4 Server

cd Hy-MT2-30B-A3B-4bit-GGUF
CUDA_VISIBLE_DEVICES=0 bash scripts/run_q4_server.sh

The script runs:

./llama.cpp/build-hyv3-cuda/bin/llama-server \
  -m Hy-MT2-30B-A3B-Q4_K_M.gguf \
  -ngl all \
  --host 127.0.0.1 \
  --port 8080

Test Translation

bash scripts/test_translation.sh

Expected style:

The weather is really nice today.

Use stop strings when serving through the OpenAI-compatible endpoint:

["<eos:6124c78e>", "<｜hy_User｜>", "<｜hy_Assistant｜>"]

Measured Local Runtime

Test machine: RTX 4090, WSL CUDA, patched llama.cpp.

Quant	GPU model buffer	Notes
Q2_K	10465.90 MiB	49/49 layers offloaded
Q3_K_M	13670.82 MiB	49/49 layers offloaded
Q4_K_M	17254.26 MiB	49/49 layers offloaded

For Q4 server translation tests after warmup, generation was typically around 195-206 tok/s on the local 4090 setup. Actual speed depends on prompt length, context size, batch settings, and build options.

Tested Cases

The Q4 server path was tested with:

normal Chinese to English translation
mixed Chinese/English/Japanese text
code and function names
Markdown lists and tables
XML tags
JSON-like text
logs, shell commands, and Windows/WSL paths

Structure-heavy prompts should explicitly ask the model to preserve structure, keys, tags, and newlines.

License

This repository contains quantized derivatives of tencent/Hy-MT2-30B-A3B. Follow the upstream model license and usage terms.

Downloads last month: 3,741

GGUF

Model size

30B params

Architecture

hy_v3

Hardware compatibility

2-bit

3-bit

4-bit

Model tree for GrahLnn/Hy-MT2-30B-A3B-4bit-GGUF

Base model

tencent/Hy-MT2-30B-A3B

Quantized

(13)

this model