Instructions to use vinnj/coachwhip-models with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use vinnj/coachwhip-models 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 vinnj/coachwhip-models
# Run inference directly in the terminal:
llama cli -hf vinnj/coachwhip-models

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf vinnj/coachwhip-models
# Run inference directly in the terminal:
llama cli -hf vinnj/coachwhip-models

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 vinnj/coachwhip-models
# Run inference directly in the terminal:
./llama-cli -hf vinnj/coachwhip-models

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 vinnj/coachwhip-models
# Run inference directly in the terminal:
./build/bin/llama-cli -hf vinnj/coachwhip-models

Use Docker

docker model run hf.co/vinnj/coachwhip-models

LM Studio
Jan

vLLM

How to use vinnj/coachwhip-models with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/vinnj/coachwhip-models

Ollama
How to use vinnj/coachwhip-models with Ollama:
```
ollama run hf.co/vinnj/coachwhip-models
```
Unsloth Desktop

How to use vinnj/coachwhip-models with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf vinnj/coachwhip-models

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": "vinnj/coachwhip-models"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use vinnj/coachwhip-models with Docker Model Runner:
```
docker model run hf.co/vinnj/coachwhip-models
```

Lemonade

How to use vinnj/coachwhip-models with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull vinnj/coachwhip-models

Run and chat with the model

lemonade run user.coachwhip-models-{{QUANT_TAG}}

List all available models

lemonade list

Hermes Agent

How to use vinnj/coachwhip-models with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf vinnj/coachwhip-models

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 vinnj/coachwhip-models

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use vinnj/coachwhip-models with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf vinnj/coachwhip-models

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 "vinnj/coachwhip-models" \
  --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"

Coachwhip models

Ready-squeezed GGUF files for Coachwhip, the engine that runs mixture-of-experts models that do not fit a 16 GB Mac by streaming the experts from the SSD. Each file here was made once with coachwhip-prepare, the tool that ships in the Coachwhip install, so you download it instead of squeezing it yourself. The engine, the install line and the start line for each model are in the Coachwhip README.

Check the hash after the download: shasum -a 256 <file>.

Files

file	bytes	sha256	squeezed from	start
`Qwen3-Coder-Next.Q2X.gguf`	29,335,336,000	`cdc0584c1e4d30134cbb83f18e4c7607af90bd1145a64940c7a357b02a0216af`	MaziyarPanahi/Qwen3-Coder-Next-GGUF Q3_K_M; gate and up experts to Q2_K, down to Q3_K, everything else untouched	`~/coachwhip/coachwhip.sh ~/Downloads/Qwen3-Coder-Next.Q2X.gguf ~/Downloads/tokenizer.json 109 8090 --parallel --flash-reads --stand-ins`

The tokenizer.json is the model's own, from Qwen/Qwen3-Coder-Next.

Licences

Each file carries the licence of the model it was squeezed from. Qwen3-Coder-Next is Apache-2.0.

Downloads last month: -

GGUF

Model size

80B params

Architecture

qwen3next

Hardware compatibility

We're not able to determine the quantization variants.

View all variants

Model tree for vinnj/coachwhip-models

Base model

Qwen/Qwen3-Coder-Next

Quantized

(117)

this model