Instructions to use OrbitMC/Mew1-2.6B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use OrbitMC/Mew1-2.6B 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 OrbitMC/Mew1-2.6B
# Run inference directly in the terminal:
llama cli -hf OrbitMC/Mew1-2.6B

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf OrbitMC/Mew1-2.6B
# Run inference directly in the terminal:
llama cli -hf OrbitMC/Mew1-2.6B

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 OrbitMC/Mew1-2.6B
# Run inference directly in the terminal:
./llama-cli -hf OrbitMC/Mew1-2.6B

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 OrbitMC/Mew1-2.6B
# Run inference directly in the terminal:
./build/bin/llama-cli -hf OrbitMC/Mew1-2.6B

Use Docker

docker model run hf.co/OrbitMC/Mew1-2.6B

LM Studio
Jan
Ollama
How to use OrbitMC/Mew1-2.6B with Ollama:
```
ollama run hf.co/OrbitMC/Mew1-2.6B
```

Unsloth Studio

How to use OrbitMC/Mew1-2.6B 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 OrbitMC/Mew1-2.6B 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 OrbitMC/Mew1-2.6B to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for OrbitMC/Mew1-2.6B to start chatting

How to use OrbitMC/Mew1-2.6B with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf OrbitMC/Mew1-2.6B

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": "OrbitMC/Mew1-2.6B"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

OpenClaw new

How to use OrbitMC/Mew1-2.6B with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf OrbitMC/Mew1-2.6B

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 "OrbitMC/Mew1-2.6B" \
  --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"

Docker Model Runner
How to use OrbitMC/Mew1-2.6B with Docker Model Runner:
```
docker model run hf.co/OrbitMC/Mew1-2.6B
```

Lemonade

How to use OrbitMC/Mew1-2.6B with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull OrbitMC/Mew1-2.6B

Run and chat with the model

lemonade run user.Mew1-2.6B-{{QUANT_TAG}}

List all available models

lemonade list

Hermes Agent

How to use OrbitMC/Mew1-2.6B with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf OrbitMC/Mew1-2.6B

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 OrbitMC/Mew1-2.6B

Run Hermes

hermes

Atomic Chat

Mew1-2.6B / README.md

OrbitMC

Upload README.md with huggingface_hub

054f375 verified about 8 hours ago

preview code

Raw

History Blame Contribute Delete

3.22 kB

	---
	library_name: gguf
	license: apache-2.0
	base_model: LiquidAI/LFM2.5-2.6B
	tags:
	- mew
	- uncensored
	- roleplay
	- lfm
	- linear-rnn
	---

	<div align="center">
	<div style="display: flex; justify-content: center; gap: 1.5em; margin-bottom: 1.5em;">
	<a href="https://huggingface.co/OrbitMC"><strong>👤 Profile</strong></a> •
	<a href="https://huggingface.co/OrbitMC/Mew1-2.6B/tree/main/ggufs"><strong>📂 All Models</strong></a>
	</div>
	</div>

	# 🐱 Mew1-2.6B

	Mew1-2.6B is a high-performance, next-generation language model developed by Unmid. Engineered for elite intelligence and uncensored freedom, Mew provides unparalleled reasoning capabilities and creative depth at a compact 2.6B parameter scale.

	Designed to push the limits of small-scale language modeling, Mew delivers instantaneous responses while maintaining a sophisticated grasp of complex instructions and nuanced roleplay. It excels in environments where low latency and high creative output are required.

	## 🌟 Key Features
	- Uncensored by Design: Optimized for unrestricted creative expression and objective reasoning.
	- Linear RNN Efficiency: Built on a specialized architecture that offers linear scaling for long contexts.
	- Elite Reasoning: Outperforms many models twice its size in logic and instruction following.

	## 📦 GGUF Downloads

	\| File \| Method \| Estimated Size \| Download \| Best For \|
	\| :--- \| :--- \| :---: \| :---: \| :--- \|
	\| Mew1-2.6B-q2-k.gguf \| Q2_K \| 0.85 GB \| [download](https://huggingface.co/OrbitMC/Mew1-2.6B/resolve/main/ggufs/Mew1-2.6B-q2-k.gguf?download=true) \| Ultra-low RAM \|
	\| Mew1-2.6B-q4-k-m.gguf \| Q4_K_M \| 1.56 GB \| [download](https://huggingface.co/OrbitMC/Mew1-2.6B/resolve/main/ggufs/Mew1-2.6B-q4-k-m.gguf?download=true) \| Balanced (Recommended) \|
	\| Mew1-2.6B-q6-k.gguf \| Q6_K \| 2.15 GB \| [download](https://huggingface.co/OrbitMC/Mew1-2.6B/resolve/main/ggufs/Mew1-2.6B-q6-k.gguf?download=true) \| High Precision \|
	\| Mew1-2.6B-q8-0.gguf \| Q8_0 \| 2.76 GB \| [download](https://huggingface.co/OrbitMC/Mew1-2.6B/resolve/main/ggufs/Mew1-2.6B-q8-0.gguf?download=true) \| Maximum Quality \|

	## 🛠️ Technical Specifications
	- Parameters: 2.6 Billion
	- Architecture: Fine-tuned Liquid Foundation Model (LFM) 2.5
	- Context Window: 32,768 Tokens
	- Training Focus: Advanced reasoning, creative writing, and technical instruction following.

	## 🎯 Use Cases
	- Creative Writing: Drafting stories, scripts, and roleplay scenarios with deep character consistency.
	- Technical Assistance: Explaining complex concepts and generating code snippets.
	- Local Deployment: Ideal for mobile devices, low-power laptops, and private server environments.

	## ⚙️ Hardware Recommendations
	\| Quantization \| Minimum RAM \| Recommended Hardware \|
	\| :--- \| :---: \| :--- \|
	\| Q2_K \| 2 GB \| Most Mobile Phones \|
	\| Q4_K_M \| 4 GB \| Standard Laptop / iPhone 13+ \|
	\| Q8_0 \| 6 GB \| Dedicated GPU / 8GB+ Desktop RAM \|

	## 🚀 Usage Instructions
	Compatible with llama.cpp, LM Studio, and KoboldCPP.

	Recommended Settings:
	- Temperature: 0.7 - 0.85 (for creativity)
	- Repeat Penalty: 1.1
	- Top-P: 0.9

	---
	Developed with ❤️ by Unmid

	---
	library_name: gguf
	license: apache-2.0
	base_model: LiquidAI/LFM2.5-2.6B
	tags:
	- mew
	- uncensored
	- roleplay
	- lfm
	- linear-rnn
	---

	<div align="center">
	<div style="display: flex; justify-content: center; gap: 1.5em; margin-bottom: 1.5em;">
	<a href="https://huggingface.co/OrbitMC"><strong>👤 Profile</strong></a> •
	<a href="https://huggingface.co/OrbitMC/Mew1-2.6B/tree/main/ggufs"><strong>📂 All Models</strong></a>
	</div>
	</div>

	# 🐱 Mew1-2.6B

	Mew1-2.6B is a high-performance, next-generation language model developed by Unmid. Engineered for elite intelligence and uncensored freedom, Mew provides unparalleled reasoning capabilities and creative depth at a compact 2.6B parameter scale.

	Designed to push the limits of small-scale language modeling, Mew delivers instantaneous responses while maintaining a sophisticated grasp of complex instructions and nuanced roleplay. It excels in environments where low latency and high creative output are required.

	## 🌟 Key Features
	- Uncensored by Design: Optimized for unrestricted creative expression and objective reasoning.
	- Linear RNN Efficiency: Built on a specialized architecture that offers linear scaling for long contexts.
	- Elite Reasoning: Outperforms many models twice its size in logic and instruction following.

	## 📦 GGUF Downloads

	\| File \| Method \| Estimated Size \| Download \| Best For \|
	\| :--- \| :--- \| :---: \| :---: \| :--- \|
	\| Mew1-2.6B-q2-k.gguf \| Q2_K \| 0.85 GB \| [download](https://huggingface.co/OrbitMC/Mew1-2.6B/resolve/main/ggufs/Mew1-2.6B-q2-k.gguf?download=true) \| Ultra-low RAM \|
	\| Mew1-2.6B-q4-k-m.gguf \| Q4_K_M \| 1.56 GB \| [download](https://huggingface.co/OrbitMC/Mew1-2.6B/resolve/main/ggufs/Mew1-2.6B-q4-k-m.gguf?download=true) \| Balanced (Recommended) \|
	\| Mew1-2.6B-q6-k.gguf \| Q6_K \| 2.15 GB \| [download](https://huggingface.co/OrbitMC/Mew1-2.6B/resolve/main/ggufs/Mew1-2.6B-q6-k.gguf?download=true) \| High Precision \|
	\| Mew1-2.6B-q8-0.gguf \| Q8_0 \| 2.76 GB \| [download](https://huggingface.co/OrbitMC/Mew1-2.6B/resolve/main/ggufs/Mew1-2.6B-q8-0.gguf?download=true) \| Maximum Quality \|

	## 🛠️ Technical Specifications
	- Parameters: 2.6 Billion
	- Architecture: Fine-tuned Liquid Foundation Model (LFM) 2.5
	- Context Window: 32,768 Tokens
	- Training Focus: Advanced reasoning, creative writing, and technical instruction following.

	## 🎯 Use Cases
	- Creative Writing: Drafting stories, scripts, and roleplay scenarios with deep character consistency.
	- Technical Assistance: Explaining complex concepts and generating code snippets.
	- Local Deployment: Ideal for mobile devices, low-power laptops, and private server environments.

	## ⚙️ Hardware Recommendations
	\| Quantization \| Minimum RAM \| Recommended Hardware \|
	\| :--- \| :---: \| :--- \|
	\| Q2_K \| 2 GB \| Most Mobile Phones \|
	\| Q4_K_M \| 4 GB \| Standard Laptop / iPhone 13+ \|
	\| Q8_0 \| 6 GB \| Dedicated GPU / 8GB+ Desktop RAM \|

	## 🚀 Usage Instructions
	Compatible with llama.cpp, LM Studio, and KoboldCPP.

	Recommended Settings:
	- Temperature: 0.7 - 0.85 (for creativity)
	- Repeat Penalty: 1.1
	- Top-P: 0.9

	---
	Developed with ❤️ by Unmid