Instructions to use continker/Qwen3.5-4B-metro-v24 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries
PEFT
How to use continker/Qwen3.5-4B-metro-v24 with PEFT:
```
Task type is invalid.
```
Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use continker/Qwen3.5-4B-metro-v24 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 continker/Qwen3.5-4B-metro-v24:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf continker/Qwen3.5-4B-metro-v24:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf continker/Qwen3.5-4B-metro-v24:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf continker/Qwen3.5-4B-metro-v24: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 continker/Qwen3.5-4B-metro-v24:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf continker/Qwen3.5-4B-metro-v24: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 continker/Qwen3.5-4B-metro-v24:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf continker/Qwen3.5-4B-metro-v24:Q4_K_M

Use Docker

docker model run hf.co/continker/Qwen3.5-4B-metro-v24:Q4_K_M

LM Studio
Jan
Ollama
How to use continker/Qwen3.5-4B-metro-v24 with Ollama:
```
ollama run hf.co/continker/Qwen3.5-4B-metro-v24:Q4_K_M
```

Unsloth Studio

How to use continker/Qwen3.5-4B-metro-v24 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 continker/Qwen3.5-4B-metro-v24 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 continker/Qwen3.5-4B-metro-v24 to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for continker/Qwen3.5-4B-metro-v24 to start chatting

How to use continker/Qwen3.5-4B-metro-v24 with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf continker/Qwen3.5-4B-metro-v24:Q4_K_M

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": "continker/Qwen3.5-4B-metro-v24:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use continker/Qwen3.5-4B-metro-v24 with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf continker/Qwen3.5-4B-metro-v24: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 continker/Qwen3.5-4B-metro-v24:Q4_K_M

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use continker/Qwen3.5-4B-metro-v24 with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf continker/Qwen3.5-4B-metro-v24: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 "continker/Qwen3.5-4B-metro-v24: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"

Docker Model Runner
How to use continker/Qwen3.5-4B-metro-v24 with Docker Model Runner:
```
docker model run hf.co/continker/Qwen3.5-4B-metro-v24:Q4_K_M
```

Lemonade

How to use continker/Qwen3.5-4B-metro-v24 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull continker/Qwen3.5-4B-metro-v24:Q4_K_M

Run and chat with the model

lemonade run user.Qwen3.5-4B-metro-v24-Q4_K_M

List all available models

lemonade list

Qwen3.5-4B-metro-v24

QLoRA fine-tune of Qwen3.5-4B for the MetroLLM-Bench transit-kiosk task. v24 is the leakage-free retraining used in the MetroLLM-Bench paper (teacher traces drawn only from the 717-case training partition; 238 cases held out). Supersedes continker/Qwen3.5-4B-metro-v23.

Held-out results (n=238, mean of 2 seeds)

Metric	4B base	4B + v24	Δ
Tier-1	89.32	91.32	+2.00
Composite	87.25	89.12	+1.87

The efficiency headline of the paper: at 2.6 GB Q4_K_M, the 4B student ties GPT-5.4 full at maximum reasoning effort on held-out tier-1 (91.32 vs 91.37).

adapter/ — LoRA adapter (rank 16, α 32; QLoRA 4-bit NF4) + tokenizer + chat template
Qwen3.5-4B-metro-v24-Q4_K_M.gguf — merged GGUF (2.6 GB)
training_summary.json

GGUF: `llama-server --hf-repo continker/Qwen3.5-4B-metro-v24 --hf-file Qwen3.5-4B-metro-v24-Q4_K_M.gguf`. The LoRA adapter keys use the `.language_model.` prefix; strip it to load onto text-only Qwen3.5-4B. Apache 2.0.