Instructions to use Infinigence/Megrez2-3x7B-A3B-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use Infinigence/Megrez2-3x7B-A3B-GGUF 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 Infinigence/Megrez2-3x7B-A3B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf Infinigence/Megrez2-3x7B-A3B-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Infinigence/Megrez2-3x7B-A3B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf Infinigence/Megrez2-3x7B-A3B-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 Infinigence/Megrez2-3x7B-A3B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf Infinigence/Megrez2-3x7B-A3B-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 Infinigence/Megrez2-3x7B-A3B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Infinigence/Megrez2-3x7B-A3B-GGUF:Q4_K_M

Use Docker

docker model run hf.co/Infinigence/Megrez2-3x7B-A3B-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use Infinigence/Megrez2-3x7B-A3B-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Infinigence/Megrez2-3x7B-A3B-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": "Infinigence/Megrez2-3x7B-A3B-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Infinigence/Megrez2-3x7B-A3B-GGUF:Q4_K_M

Ollama
How to use Infinigence/Megrez2-3x7B-A3B-GGUF with Ollama:
```
ollama run hf.co/Infinigence/Megrez2-3x7B-A3B-GGUF:Q4_K_M
```
Unsloth Desktop
Docker Model Runner
How to use Infinigence/Megrez2-3x7B-A3B-GGUF with Docker Model Runner:
```
docker model run hf.co/Infinigence/Megrez2-3x7B-A3B-GGUF:Q4_K_M
```

Lemonade

How to use Infinigence/Megrez2-3x7B-A3B-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Infinigence/Megrez2-3x7B-A3B-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Megrez2-3x7B-A3B-GGUF-Q4_K_M

List all available models

lemonade list

Atomic Chat

Megrez2-3x7B-A3B

🔗 Github | 📄 Tech Report | 💻 Demo | 💬 WeChat Official

中文 | English

Introduction

Megrez2-3x7B-A3B is a device native large language model. Megrez2 takes advantages of both the accuracy of Mixture-of-Experts (MoE) architecture and the compact size of Dense models. This release model was trained on 8T Tokens of data. In the future, we plan to improve the model's reasoning and agent capabilities.

Model Card


Architecture	Mixture-of-Experts (MoE)
Total Parameters	3x7B
Activated Parameters	3B
Experts Shared Frequency	3
Number of Layers (Dense layer included)	31
Number of Dense Layers	1
Attention Hidden Dimension	2048
MoE Hidden Dimension (per Expert)	1408
Number of Attention Heads	16
Number of Experts	64
Selected Experts per Token	6
Number of Shared Experts	4
Vocabulary Size	128,880
Context Length	32K
Base Frequency of RoPE	5,000,000
Attention Mechanism	GQA
Activation Function	SwiGLU

Performance

We evaluated Megrez2-3x7B-A3B using the open-source evaluation tool OpenCompass on several important benchmarks. Some of the evaluation results are shown in the table below.

Benchmark	Metric	^{Megrez2-3x7B -A3B}	^{Megrez2-3x7B -A3B-Preview}	^{SmallThinker-21B -A3B-Instruct}	^{Qwen3-30B-A3B}	^Qwen3-8B	^{Qwen3-4B -Instruct-2507}	^{Phi4-14B (nothink)}	^Gemma3-12B
Activate Params (B)		3.0	3.0	3.0	3.3	8.2	4.0	14.7	12.2
Stored Params (B)		7.5	7.5	21.5	30.5	8.2	4.0	14.7	12.2
MMLU	EM	85.4	87.5	84.4	85.1	81.8	-	84.6	78.5
GPQA	EM	58.8	28.8	55.0	44.4	38.9	62	55.5	34.9
IFEval	Inst loose	87.7	80.2	85.8	84.3	83.9	83.4	63.2	74.7
MATH-500	EM	87.2	81.6	82.4	84.4	81.6	-	80.2	82.4

How to Run

llama.cpp

llama.cpp enables LLM inference with minimal setup and state-of-the-art performance on a wide range of hardware. Now supported, please refer to the support-megrez branch for details.

Under the FP16 floating-point precision configuration, the performance of the current model on code tasks has decreased compared to the original model. We have launched optimization efforts to address this issue and are currently exploring solutions.

Best Practice

To achieve optimal performance, we recommend the following settings:

Sampling Parameters: we suggest using Temperature=0.7 and TopP=0.9 .
Standardize Output Format: We recommend using prompts to standardize model outputs when benchmarking.
- Math Problems: Include "Please reason step by step, and put your final answer within \boxed{}." in the prompt.
- Multiple-Choice Questions: Add the following JSON structure to the prompt to standardize responses: "Please show your choice in the answer field with only the choice letter, e.g., "answer": "C"."

License Agreement

All our open-weight models are licensed under Apache 2.0.

Citation

If you find our work helpful, feel free to give us a cite.

@misc{li2025megrez2technicalreport,
      title={Megrez2 Technical Report}, 
      author={Boxun Li and Yadong Li and Zhiyuan Li and Congyi Liu and Weilin Liu and Guowei Niu and Zheyue Tan and Haiyang Xu and Zhuyu Yao and Tao Yuan and Dong Zhou and Yueqing Zhuang and Bo Zhao and Guohao Dai and Yu Wang},
      year={2025},
      eprint={2507.17728},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2507.17728}, 
}

Contact

If you have any questions, please feel free to submit a GitHub issue or contact WeChat groups.

Downloads last month: 522

GGUF

Model size

7B params

Architecture

megrez-moe

Hardware compatibility

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

16-bit

Collection including Infinigence/Megrez2-3x7B-A3B-GGUF

Megrez2

Collection

6 items • Updated Sep 15, 2025 • 1

Paper for Infinigence/Megrez2-3x7B-A3B-GGUF

Megrez2 Technical Report

Paper • 2507.17728 • Published Jul 23, 2025