Instructions to use ahpah/YokoCoder-4B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use ahpah/YokoCoder-4B 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 ahpah/YokoCoder-4B:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf ahpah/YokoCoder-4B:Q4_K_M

Install from WinGet (Windows)

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

Use Docker

docker model run hf.co/ahpah/YokoCoder-4B:Q4_K_M

Llama new
LM Studio
Jan

vLLM

How to use ahpah/YokoCoder-4B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ahpah/YokoCoder-4B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ahpah/YokoCoder-4B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ahpah/YokoCoder-4B:Q4_K_M

Ollama
How to use ahpah/YokoCoder-4B with Ollama:
```
ollama run hf.co/ahpah/YokoCoder-4B:Q4_K_M
```
Unsloth Desktop

How to use ahpah/YokoCoder-4B with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf ahpah/YokoCoder-4B: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": "ahpah/YokoCoder-4B:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use ahpah/YokoCoder-4B with Docker Model Runner:
```
docker model run hf.co/ahpah/YokoCoder-4B:Q4_K_M
```

Lemonade

How to use ahpah/YokoCoder-4B with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull ahpah/YokoCoder-4B:Q4_K_M

Run and chat with the model

lemonade run user.YokoCoder-4B-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use ahpah/YokoCoder-4B with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf ahpah/YokoCoder-4B: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 ahpah/YokoCoder-4B:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use ahpah/YokoCoder-4B with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf ahpah/YokoCoder-4B: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 "ahpah/YokoCoder-4B: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"

YokoCoder-4B / README.md

ahpah

Document verified Ollama generation for the five-epoch release

63e6d5b verified 11 days ago

preview code

Raw History Blame Contribute Delete

2.44 kB

---
license: apache-2.0
base_model: wangzhang/Qwen3.5-4B-abliterated
pipeline_tag: text-generation
library_name: gguf
tags:
- gguf
- qwen3.5
- code
- coding
- abliterated
- ollama
language:
- en
---

# YokoCoder-4B

YokoCoder-4B is a compact, locally runnable coding assistant built on the abliterated Qwen3.5 4B model. It was fine-tuned for coding, further trained on Flappy Bird examples, and continued for five more epochs on a new coding dataset. It builds on the author's checkpoint that passed the Flappy Bird game-generation test. This five-epoch release was tested in Ollama and passed a three-case Python function smoke test, offering practical coding value for its size.

## Model details

- **Format:** GGUF, Q4_K_M
- **Parameters:** 4B
- **Download size:** about 2.7 GB
- **Base:** [wangzhang/Qwen3.5-4B-abliterated](https://huggingface.co/wangzhang/Qwen3.5-4B-abliterated), derived from Qwen3.5-4B
- **Training:** coding fine-tune, three additional epochs on Flappy Bird examples, then five additional epochs on 12 coding examples (28 training windows)
- **License:** Apache 2.0

## Run with Ollama

Download the files in this repository, then run the following from their directory:

```bash
ollama create yokocoder-4b -f Modelfile
ollama run yokocoder-4b
```

The included Modelfile configures a 16,384-token context and sets `num_predict -1` for generation without a fixed output-token cap. The GGUF can also be used with other compatible Qwen3.5 GGUF runtimes.

## Attribution

YokoCoder-4B is a fine-tuned and quantized derivative of [Qwen3.5-4B-abliterated](https://huggingface.co/wangzhang/Qwen3.5-4B-abliterated) by Wangzhang Wu, based on [Qwen3.5-4B](https://huggingface.co/Qwen/Qwen3.5-4B). The original model card and notices are included in [UPSTREAM_MODEL_CARD.md](UPSTREAM_MODEL_CARD.md), and the license is in [LICENSE](LICENSE).

## Latest training update

The October 29, 2026 update continues the existing YokoCoder checkpoint for five epochs, completing 20 optimizer steps with training loss 1.0559. The Q4_K_M file contains the merged updated weights. The continued-training LoRA adapter and its training records are included under `adapters/dataset4-5epochs/`. Export hashes and provenance are in `export_manifest_5epochs.json`.

The local generation check and model identity are recorded in [ollama_smoke_test_5epochs.json](ollama_smoke_test_5epochs.json).

	---
	license: apache-2.0
	base_model: wangzhang/Qwen3.5-4B-abliterated
	pipeline_tag: text-generation
	library_name: gguf
	tags:
	- gguf
	- qwen3.5
	- code
	- coding
	- abliterated
	- ollama
	language:
	- en
	---

	# YokoCoder-4B

	YokoCoder-4B is a compact, locally runnable coding assistant built on the abliterated Qwen3.5 4B model. It was fine-tuned for coding, further trained on Flappy Bird examples, and continued for five more epochs on a new coding dataset. It builds on the author's checkpoint that passed the Flappy Bird game-generation test. This five-epoch release was tested in Ollama and passed a three-case Python function smoke test, offering practical coding value for its size.

	## Model details

	- Format: GGUF, Q4_K_M
	- Parameters: 4B
	- Download size: about 2.7 GB
	- Base: [wangzhang/Qwen3.5-4B-abliterated](https://huggingface.co/wangzhang/Qwen3.5-4B-abliterated), derived from Qwen3.5-4B
	- Training: coding fine-tune, three additional epochs on Flappy Bird examples, then five additional epochs on 12 coding examples (28 training windows)
	- License: Apache 2.0

	## Run with Ollama

	Download the files in this repository, then run the following from their directory:

	```bash
	ollama create yokocoder-4b -f Modelfile
	ollama run yokocoder-4b
	```

	The included Modelfile configures a 16,384-token context and sets `num_predict -1` for generation without a fixed output-token cap. The GGUF can also be used with other compatible Qwen3.5 GGUF runtimes.

	## Attribution

	YokoCoder-4B is a fine-tuned and quantized derivative of [Qwen3.5-4B-abliterated](https://huggingface.co/wangzhang/Qwen3.5-4B-abliterated) by Wangzhang Wu, based on [Qwen3.5-4B](https://huggingface.co/Qwen/Qwen3.5-4B). The original model card and notices are included in [UPSTREAM_MODEL_CARD.md](UPSTREAM_MODEL_CARD.md), and the license is in [LICENSE](LICENSE).

	## Latest training update

	The October 29, 2026 update continues the existing YokoCoder checkpoint for five epochs, completing 20 optimizer steps with training loss 1.0559. The Q4_K_M file contains the merged updated weights. The continued-training LoRA adapter and its training records are included under `adapters/dataset4-5epochs/`. Export hashes and provenance are in `export_manifest_5epochs.json`.

	The local generation check and model identity are recorded in [ollama_smoke_test_5epochs.json](ollama_smoke_test_5epochs.json).