Instructions to use cipher-000/edge0-8b-a1b-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 cipher-000/edge0-8b-a1b-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 cipher-000/edge0-8b-a1b-gguf
# Run inference directly in the terminal:
llama cli -hf cipher-000/edge0-8b-a1b-gguf

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf cipher-000/edge0-8b-a1b-gguf
# Run inference directly in the terminal:
llama cli -hf cipher-000/edge0-8b-a1b-gguf

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 cipher-000/edge0-8b-a1b-gguf
# Run inference directly in the terminal:
./llama-cli -hf cipher-000/edge0-8b-a1b-gguf

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 cipher-000/edge0-8b-a1b-gguf
# Run inference directly in the terminal:
./build/bin/llama-cli -hf cipher-000/edge0-8b-a1b-gguf

Use Docker

docker model run hf.co/cipher-000/edge0-8b-a1b-gguf

Llama new
LM Studio
Jan

vLLM

How to use cipher-000/edge0-8b-a1b-gguf with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "cipher-000/edge0-8b-a1b-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": "cipher-000/edge0-8b-a1b-gguf",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/cipher-000/edge0-8b-a1b-gguf

Ollama
How to use cipher-000/edge0-8b-a1b-gguf with Ollama:
```
ollama run hf.co/cipher-000/edge0-8b-a1b-gguf
```
Unsloth Desktop

How to use cipher-000/edge0-8b-a1b-gguf with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf cipher-000/edge0-8b-a1b-gguf

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "cipher-000/edge0-8b-a1b-gguf"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use cipher-000/edge0-8b-a1b-gguf with Docker Model Runner:
```
docker model run hf.co/cipher-000/edge0-8b-a1b-gguf
```

Lemonade

How to use cipher-000/edge0-8b-a1b-gguf with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull cipher-000/edge0-8b-a1b-gguf

Run and chat with the model

lemonade run user.edge0-8b-a1b-gguf-{{QUANT_TAG}}

List all available models

lemonade list

Hermes Agent

How to use cipher-000/edge0-8b-a1b-gguf with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf cipher-000/edge0-8b-a1b-gguf

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 cipher-000/edge0-8b-a1b-gguf

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use cipher-000/edge0-8b-a1b-gguf with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf cipher-000/edge0-8b-a1b-gguf

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 "cipher-000/edge0-8b-a1b-gguf" \
  --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"

Edge0-8B-A1B-preview - official-pipeline GGUF (bailingmoe3)

GGUF build of Edge0/Edge0-8B-A1B-preview, produced with Edge0's own converter (windows/tools/convert_mlx_to_gguf.py) from the official MLX checkpoint - not a third-party requantisation.

Why this build

Community GGUF conversions of this model exist, but they merge the Recover-LoRA into the base weights, requantise afterwards, and discard the prerouter tensors. This build keeps the official converter's structure instead:

canonical bailingmoe3 tensor naming
Recover-LoRA kept as a separate adapter (lora_edge0_8b-gguf.gguf), never merged
prerouter/pregate sidecar tensors preserved rather than stripped
the converter's own gates run and recorded (tensor-count identity, sentinel sha, kv_b transform reversibility spot-check, exp dual-path)

The conversion is byte-reproducible; manifest.json records input and artefact sha256s.

Files

File	Purpose
`edge0-8b.gguf`	base model
`lora_edge0_8b-gguf.gguf`	Recover-LoRA adapter (loads alongside the base)
`manifest.json`	provenance, sizes, sha256s, converter revision

Architecture

general.architecture = bailingmoe3 - 24 layers, 128 experts (8 active), 1 shared expert, vocab 157184. Hybrid MLA + KDA linear attention.

For the Edge0 Android app, import edge0-8b.gguf via Models -> picker; the runtime picks up the lora_ companion automatically.

Attribution

Original model, framework and weights: Edge0-AI/Edge0 (Apache-2.0). Base lineage: inclusionAI/Ling-3.0-tiny-base. This repository is an unaffiliated repack and is not an official Edge0 release.

Downloads last month: 526

GGUF

Model size

8B params

Architecture

bailingmoe3

Hardware compatibility

We're not able to determine the quantization variants.

View all variants

Model tree for cipher-000/edge0-8b-a1b-gguf

Base model

inclusionAI/Ling-3.0-tiny-base

Adapter

Edge0/Edge0-8B-A1B-preview

Quantized

(2)

this model