Instructions to use Doctorgp1/sabi-v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use Doctorgp1/sabi-v1 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 Doctorgp1/sabi-v1:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf Doctorgp1/sabi-v1:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Doctorgp1/sabi-v1:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf Doctorgp1/sabi-v1: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 Doctorgp1/sabi-v1:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf Doctorgp1/sabi-v1: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 Doctorgp1/sabi-v1:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Doctorgp1/sabi-v1:Q4_K_M

Use Docker

docker model run hf.co/Doctorgp1/sabi-v1:Q4_K_M

LM Studio
Jan
Ollama
How to use Doctorgp1/sabi-v1 with Ollama:
```
ollama run hf.co/Doctorgp1/sabi-v1:Q4_K_M
```
Unsloth Desktop

How to use Doctorgp1/sabi-v1 with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Doctorgp1/sabi-v1:Q4_K_M

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": "Doctorgp1/sabi-v1:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use Doctorgp1/sabi-v1 with Docker Model Runner:
```
docker model run hf.co/Doctorgp1/sabi-v1:Q4_K_M
```

Lemonade

How to use Doctorgp1/sabi-v1 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Doctorgp1/sabi-v1:Q4_K_M

Run and chat with the model

lemonade run user.sabi-v1-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use Doctorgp1/sabi-v1 with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Doctorgp1/sabi-v1:Q4_K_M

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 Doctorgp1/sabi-v1:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use Doctorgp1/sabi-v1 with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Doctorgp1/sabi-v1:Q4_K_M

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 "Doctorgp1/sabi-v1:Q4_K_M" \
  --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"

sabi-v1 (Qwen2.5-Coder-3B-Instruct, Q4_K_M GGUF)

This is an unmodified mirror of Qwen's official Qwen/Qwen2.5-Coder-3B-Instruct-GGUF release (qwen2.5-coder-3b-instruct-q4_k_m.gguf), re-hosted here for convenience as the default model used by SABI, a private, fully offline AI coworker that runs on a standard 8 GB laptop under a 7 GB memory budget. No weights, architecture, or quantization were changed — this is the same file, saved under SABI's local model name (sabi-v1.Q4_K_M.gguf).

Built with Qwen.

SHA256: 724fb256bec1ff062b2f65e4569e871ad2e95ab2a3989723d1769c54294730b7
Size: 2,104,932,800 bytes

All credit for the model itself goes to the Qwen team at Alibaba Cloud. For the canonical release, full model card, and other quantizations, see the original repository.

License

Distributed under the Qwen RESEARCH LICENSE AGREEMENT (non-commercial — research/evaluation use only; see LICENSE and NOTICE in this repo, and the original license). This is not covered by SABI's own MIT license — SABI's application code is MIT, but this underlying model is not.

In SABI

SABI downloads this file automatically via sabi download / download_model.sh, which by default pulls it straight from Qwen's official repo — this mirror is provided purely for visibility/convenience, not as a required dependency.