Instructions to use dipeshmajithia/Mirror-80M-MoE-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use dipeshmajithia/Mirror-80M-MoE-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="dipeshmajithia/Mirror-80M-MoE-GGUF",
	filename="mirror_ai_chat_v1.gguf",
)

output = llm(
	"Once upon a time,",
	max_tokens=512,
	echo=True
)
print(output)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use dipeshmajithia/Mirror-80M-MoE-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 dipeshmajithia/Mirror-80M-MoE-GGUF:F32
# Run inference directly in the terminal:
llama cli -hf dipeshmajithia/Mirror-80M-MoE-GGUF:F32

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf dipeshmajithia/Mirror-80M-MoE-GGUF:F32
# Run inference directly in the terminal:
llama cli -hf dipeshmajithia/Mirror-80M-MoE-GGUF:F32

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 dipeshmajithia/Mirror-80M-MoE-GGUF:F32
# Run inference directly in the terminal:
./llama-cli -hf dipeshmajithia/Mirror-80M-MoE-GGUF:F32

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 dipeshmajithia/Mirror-80M-MoE-GGUF:F32
# Run inference directly in the terminal:
./build/bin/llama-cli -hf dipeshmajithia/Mirror-80M-MoE-GGUF:F32

Use Docker

docker model run hf.co/dipeshmajithia/Mirror-80M-MoE-GGUF:F32

LM Studio
Jan

vLLM

How to use dipeshmajithia/Mirror-80M-MoE-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "dipeshmajithia/Mirror-80M-MoE-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "dipeshmajithia/Mirror-80M-MoE-GGUF",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/dipeshmajithia/Mirror-80M-MoE-GGUF:F32

Ollama
How to use dipeshmajithia/Mirror-80M-MoE-GGUF with Ollama:
```
ollama run hf.co/dipeshmajithia/Mirror-80M-MoE-GGUF:F32
```

Unsloth Studio

How to use dipeshmajithia/Mirror-80M-MoE-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 dipeshmajithia/Mirror-80M-MoE-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 dipeshmajithia/Mirror-80M-MoE-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for dipeshmajithia/Mirror-80M-MoE-GGUF to start chatting

Atomic Chat new
Docker Model Runner
How to use dipeshmajithia/Mirror-80M-MoE-GGUF with Docker Model Runner:
```
docker model run hf.co/dipeshmajithia/Mirror-80M-MoE-GGUF:F32
```

Lemonade

How to use dipeshmajithia/Mirror-80M-MoE-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull dipeshmajithia/Mirror-80M-MoE-GGUF:F32

Run and chat with the model

lemonade run user.Mirror-80M-MoE-GGUF-F32

List all available models

lemonade list

You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

Mirror-MoE-80M-GGUF

The GGUF format quantization of Mirror-MoE-80M, a Sparse Mixture-of-Experts language model optimized for edge devices.

Original PyTorch/MLX Model

📦 Available Models

We offer two distinct variants of Mirror-MoE in various quantization levels:

1. Hybrid (Chat & Retrieval)

Best for general conversation, fact retrieval, and casual interactions.

mirror_moe_hybrid_f32.gguf - Full precision (32-bit).
mirror_moe_hybrid_q8_0.gguf - High quality 8-bit quantization.
mirror_moe_hybrid_q4_k.gguf - Balanced 4-bit quantization (Recommended for mobile).

2. Elite (Reasoning & Logic)

Best for logic puzzles, math, coding, and strict instruction following.

mirror_moe_elite_f32.gguf - Full precision (32-bit).
mirror_moe_elite_q8_0.gguf - High quality 8-bit quantization.
mirror_moe_elite_q4_k.gguf - Balanced 4-bit quantization (Recommended for mobile).

🚀 Model Specs

Metric	Value
Total Parameters	81M
Active Parameters	37M (2.2x sparse)
Experts	16 Sparse + 1 Shared Anchor
Context Window	512 tokens

💻 Usage

These files are compatible with llama.cpp and other GGUF-supporting libraries.

CLI Example

./main -m mirror_moe_hybrid_q4_k.gguf -p "Hello, how are you?" -n 128

📜 Citation

@misc{mirror2026moe,
  title={Mirror-MoE-80M: Anchor-Stabilized Granular Mixture of Experts for Low-Resource Training},
  author={Dipesh Majithia},
  year={2026},
  publisher={Zenodo},
  doi={10.5281/zenodo.18473273},
  url={https://zenodo.org/records/18473273}
}

📄 License

CC BY 4.0 - Free to use with attribution to MirrorAI / Dipesh Majithia.

Downloads last month: 2

GGUF

Model size

81.1M params

Architecture

mirror_ai

Hardware compatibility

8-bit

32-bit

View +1 variant