Instructions to use ZenithLLM/zen-alta-draft-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 ZenithLLM/zen-alta-draft-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 ZenithLLM/zen-alta-draft-gguf:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf ZenithLLM/zen-alta-draft-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 ZenithLLM/zen-alta-draft-gguf:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf ZenithLLM/zen-alta-draft-gguf: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 ZenithLLM/zen-alta-draft-gguf:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf ZenithLLM/zen-alta-draft-gguf: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 ZenithLLM/zen-alta-draft-gguf:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf ZenithLLM/zen-alta-draft-gguf:Q4_K_M

Use Docker

docker model run hf.co/ZenithLLM/zen-alta-draft-gguf:Q4_K_M

LM Studio
Jan

vLLM

How to use ZenithLLM/zen-alta-draft-gguf with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/ZenithLLM/zen-alta-draft-gguf:Q4_K_M

Ollama
How to use ZenithLLM/zen-alta-draft-gguf with Ollama:
```
ollama run hf.co/ZenithLLM/zen-alta-draft-gguf:Q4_K_M
```
Unsloth Desktop

How to use ZenithLLM/zen-alta-draft-gguf with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf ZenithLLM/zen-alta-draft-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": "ZenithLLM/zen-alta-draft-gguf:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use ZenithLLM/zen-alta-draft-gguf with Docker Model Runner:
```
docker model run hf.co/ZenithLLM/zen-alta-draft-gguf:Q4_K_M
```

Lemonade

How to use ZenithLLM/zen-alta-draft-gguf with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull ZenithLLM/zen-alta-draft-gguf:Q4_K_M

Run and chat with the model

lemonade run user.zen-alta-draft-gguf-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use ZenithLLM/zen-alta-draft-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 ZenithLLM/zen-alta-draft-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 ZenithLLM/zen-alta-draft-gguf:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use ZenithLLM/zen-alta-draft-gguf with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf ZenithLLM/zen-alta-draft-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 "ZenithLLM/zen-alta-draft-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"

⚡ Zen Alta Draft (GGUF Quantized)

This repository provides the quantized Q4_K_M GGUF binary (545.72 MB) for the Zen Alta 4-Layer Speculative Decoding Draft Model.

Consolidated Repository: The full model with both FP16 Safetensors weights and GGUF quantization is available at ZenithLLM/ZenAlta-Draft.

🚀 Speculative Decoding Usage in llama.cpp

Pair this draft model with the target model ZenithLLM/ZenAlta-1-3B-Phase2-GGUF:

./llama-cli \
  -m ZenAlta-1-3B-Pruned.Q4_K_M.gguf \
  -md zen-alta-draft-q4_k_m.gguf \
  --draft-max 5 \
  -p "<|start_header_id|>user<|end_header_id|>\n\nhey what are you doing<|eot_id|><|start_header_id|>assistant<|end_header_id|>\n\n" \
  -n 128

📦 File Information

File: zen-alta-draft-q4_k_m.gguf
Size: 545.72 MB
Quantization: Q4_K_M (5.67 bits per weight)
Vocabulary: 128,256 BPE (Identical to Llama 3.2 3B & Zen Alta)

Downloads last month: 101

GGUF

Model size

0.8B params

Architecture

llama

Hardware compatibility

4-bit

Model tree for ZenithLLM/zen-alta-draft-gguf

Base model

ZenithLLM/ZenAlta-1-3B-Pruned

Adapter

ZenithLLM/ZenAlta-1-3B-Phase2

Quantized

(3)

this model