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

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use cay-ai/cay 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 cay-ai/cay:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf cay-ai/cay:Q4_K_M

Install from WinGet (Windows)

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

Use Docker

docker model run hf.co/cay-ai/cay:Q4_K_M

Llama new
LM Studio
Jan
Ollama
How to use cay-ai/cay with Ollama:
```
ollama run hf.co/cay-ai/cay:Q4_K_M
```
Unsloth Desktop

How to use cay-ai/cay with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf cay-ai/cay: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": "cay-ai/cay:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use cay-ai/cay with Docker Model Runner:
```
docker model run hf.co/cay-ai/cay:Q4_K_M
```

Lemonade

How to use cay-ai/cay with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull cay-ai/cay:Q4_K_M

Run and chat with the model

lemonade run user.cay-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use cay-ai/cay with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf cay-ai/cay: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 cay-ai/cay:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use cay-ai/cay with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf cay-ai/cay: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 "cay-ai/cay: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"

Cay models

The files Cay, a local personal agent for the laptop, downloads on first run. cay doctor --fetch-models checks each one against a hash built into the app.

File	What it is	Licence
`cay-router-1.7b-Q4_K_M.gguf`	Qwen3-1.7B with Cay's router fine-tune merged in	Apache-2.0
`cay-router-0.6b-Q4_K_M.gguf`	Qwen3-0.6B with the same fine-tune merged in	Apache-2.0
`indicxlit/`	IndicXlit exported to ONNX (int8), for romanising Hindi words	MIT
`srota-0.6b-hinglish-188ae2b6-Q8_0.gguf`, `mmproj-srota-...`	Srota (moorlee/qwen3-asr-0.6b-hinglish) as published, converted to GGUF, for speech	Apache-2.0, as its model card states

The routers are trained only on synthetic commands. Base models by the Qwen team; IndicXlit by AI4Bharat. Srota is by its authors, and its training data includes OpenSLR-104 (CC BY-SA 4.0).

Downloads last month: 109

GGUF

Model size

0.6B params

Architecture

qwen3

Hardware compatibility

4-bit

8-bit

Inference Providers NEW

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