Instructions to use zonay/qwen3-bash-v3-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 zonay/qwen3-bash-v3-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 zonay/qwen3-bash-v3-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf zonay/qwen3-bash-v3-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf zonay/qwen3-bash-v3-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf zonay/qwen3-bash-v3-GGUF:Q4_K_M

Build from source code

git clone https://github.com/ggml-org/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-app
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama serve -hf zonay/qwen3-bash-v3-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama cli -hf zonay/qwen3-bash-v3-GGUF:Q4_K_M

Use Docker

docker model run hf.co/zonay/qwen3-bash-v3-GGUF:Q4_K_M

Llama new
LM Studio
Jan
Ollama
How to use zonay/qwen3-bash-v3-GGUF with Ollama:
```
ollama run hf.co/zonay/qwen3-bash-v3-GGUF:Q4_K_M
```
Unsloth Desktop

How to use zonay/qwen3-bash-v3-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf zonay/qwen3-bash-v3-GGUF: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": "zonay/qwen3-bash-v3-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use zonay/qwen3-bash-v3-GGUF with Docker Model Runner:
```
docker model run hf.co/zonay/qwen3-bash-v3-GGUF:Q4_K_M
```

Lemonade

How to use zonay/qwen3-bash-v3-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull zonay/qwen3-bash-v3-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.qwen3-bash-v3-GGUF-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use zonay/qwen3-bash-v3-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 zonay/qwen3-bash-v3-GGUF: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 zonay/qwen3-bash-v3-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use zonay/qwen3-bash-v3-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf zonay/qwen3-bash-v3-GGUF: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 "zonay/qwen3-bash-v3-GGUF: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"

qwen3-bash-v3-GGUF

Quantized GGUF of zonay/qwen3-bash-v3 — a 0.6B English → Bash translator. Runs on any CPU (Windows / Linux / Mac).

Files

file	size	use
`qwen3-bash-v3-q4_k_m.gguf`	378MB	recommended, best quality/size

Scores (held-out, first-command match + `bash -n`)

Measured on the MLX source model; Q4 typically costs 1-2 points:

big eval 500: 482/500 (96%)
broad 36 hand prompts: 34/36

Training data: zonay/bash-data-nl2sh (140k train + 2k eval synthetic pairs) — shell core, networking (ssh/dns/NAT), AI tooling (ollama/lms/opencode/llamacpp/mlx/hf/vllm), macOS dev (xcode), fullstack (js/py/db/devops), lab-authorized destructive+pentest.

Prompt tip: keep temperature low (0.0-0.2) and ask for one command.

Use cases

CLI copilot: ask in English, get the command, confirm, run.
Learning aid: inspect generated commands to learn bash idioms.
Ops runbooks: draft incident snippets from plain English, review before running.

Limitations

0.6B: strong on trained patterns, weak on unseen flags. Rare verbs may be confused.
Outputs are single commands, not multi-step plans.
risk: destructive style outputs must be confirmed by a human.

Safety

Never auto-execute outputs. Run destructive commands only on disposable lab VMs you own. Test in a sandbox first (docker run --rm --network none, timeout).

Downloads last month: 154

GGUF

Model size

0.6B params

Architecture

qwen3

Hardware compatibility

4-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for zonay/qwen3-bash-v3-GGUF

Base model

Qwen/Qwen3-0.6B-Base

Finetuned

Qwen/Qwen3-0.6B

Adapter

zonay/qwen3-bash-v3

Quantized

(1)

this model

qwen3-bash-v3-GGUF

Files

Scores (held-out, first-command match + bash -n)

Use cases

Limitations

Safety

Model tree for zonay/qwen3-bash-v3-GGUF

Scores (held-out, first-command match + `bash -n`)