Instructions to use GEHIRNS/magi-alpha-2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use GEHIRNS/magi-alpha-2 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 GEHIRNS/magi-alpha-2
# Run inference directly in the terminal:
llama cli -hf GEHIRNS/magi-alpha-2

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf GEHIRNS/magi-alpha-2
# Run inference directly in the terminal:
llama cli -hf GEHIRNS/magi-alpha-2

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 GEHIRNS/magi-alpha-2
# Run inference directly in the terminal:
./llama-cli -hf GEHIRNS/magi-alpha-2

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 GEHIRNS/magi-alpha-2
# Run inference directly in the terminal:
./build/bin/llama-cli -hf GEHIRNS/magi-alpha-2

Use Docker

docker model run hf.co/GEHIRNS/magi-alpha-2

Llama new
LM Studio
Jan
Ollama
How to use GEHIRNS/magi-alpha-2 with Ollama:
```
ollama run hf.co/GEHIRNS/magi-alpha-2
```
Unsloth Desktop

How to use GEHIRNS/magi-alpha-2 with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf GEHIRNS/magi-alpha-2

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": "GEHIRNS/magi-alpha-2"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use GEHIRNS/magi-alpha-2 with Docker Model Runner:
```
docker model run hf.co/GEHIRNS/magi-alpha-2
```

Lemonade

How to use GEHIRNS/magi-alpha-2 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull GEHIRNS/magi-alpha-2

Run and chat with the model

lemonade run user.magi-alpha-2-{{QUANT_TAG}}

List all available models

lemonade list

Hermes Agent

How to use GEHIRNS/magi-alpha-2 with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf GEHIRNS/magi-alpha-2

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 GEHIRNS/magi-alpha-2

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use GEHIRNS/magi-alpha-2 with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf GEHIRNS/magi-alpha-2

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 "GEHIRNS/magi-alpha-2" \
  --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"

NOT RECOMMENDED: experimental build. Use magi-1.0 instead.

Not recommended. This is an experimental build. Use magi-1.0 instead.

Website: gehirns.pages.dev

magi-alpha-2 (retired experimental release)

Retired, not recommended. This supervised fine-tune did not improve on the untrained base: LiveCodeBench 48/150 against 50/150 for base Qwen3.5-9B Q4_K_M under the same think-enabled protocol, a difference within run-to-run variation (five runs of the same base scored 42 to 50). It is kept online for transparency. For a small, efficient model, see magi-1.0: no training, with 96% of BF16's benchmark score in 30% of the GPU memory.

9B coder build: Qwen3.5-9B base + SFT teaching diet v2, crushed to Q4_K_M (5.24 GB). From the GEHIRNS density program — maximum capability per gigabyte. Alpha quality: experimental, kept for the record.

LiveCodeBench: 48/150 (think-enabled, greedy single-pass, measured 2026-10-16)
Base: Qwen3.5-9B (Unsloth GGUF) + LoRA diet v2 (613 rows), merged → GGUF → Q4_K_M
Context: 32768 (see Modelfile)
Lineage: ex-seele9-d2; registry alias magi-alpha-2-q4km

Run (Ollama)

ollama create magi-alpha-2-q4km -f Modelfile
ollama run magi-alpha-2-q4km

Honesty section

48/150 (32%) is a dead heat with magi-alpha-1's 49/150: diet v2 moved nothing. The supervised fine-tuning line was stopped on 2026-10-18, after a postmortem found mismatches between the training and evaluation protocols; a third diet was trained but not released. This artifact records the plateau point.