Instructions to use 0xzknw/Granite-4.2-3B-Heretic-NX-PRIME-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 0xzknw/Granite-4.2-3B-Heretic-NX-PRIME-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 0xzknw/Granite-4.2-3B-Heretic-NX-PRIME-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf 0xzknw/Granite-4.2-3B-Heretic-NX-PRIME-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 0xzknw/Granite-4.2-3B-Heretic-NX-PRIME-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf 0xzknw/Granite-4.2-3B-Heretic-NX-PRIME-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 0xzknw/Granite-4.2-3B-Heretic-NX-PRIME-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf 0xzknw/Granite-4.2-3B-Heretic-NX-PRIME-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 0xzknw/Granite-4.2-3B-Heretic-NX-PRIME-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf 0xzknw/Granite-4.2-3B-Heretic-NX-PRIME-GGUF:Q4_K_M

Use Docker

docker model run hf.co/0xzknw/Granite-4.2-3B-Heretic-NX-PRIME-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use 0xzknw/Granite-4.2-3B-Heretic-NX-PRIME-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "0xzknw/Granite-4.2-3B-Heretic-NX-PRIME-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": "0xzknw/Granite-4.2-3B-Heretic-NX-PRIME-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/0xzknw/Granite-4.2-3B-Heretic-NX-PRIME-GGUF:Q4_K_M

Ollama
How to use 0xzknw/Granite-4.2-3B-Heretic-NX-PRIME-GGUF with Ollama:
```
ollama run hf.co/0xzknw/Granite-4.2-3B-Heretic-NX-PRIME-GGUF:Q4_K_M
```
Unsloth Desktop

How to use 0xzknw/Granite-4.2-3B-Heretic-NX-PRIME-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf 0xzknw/Granite-4.2-3B-Heretic-NX-PRIME-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": "0xzknw/Granite-4.2-3B-Heretic-NX-PRIME-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use 0xzknw/Granite-4.2-3B-Heretic-NX-PRIME-GGUF with Docker Model Runner:
```
docker model run hf.co/0xzknw/Granite-4.2-3B-Heretic-NX-PRIME-GGUF:Q4_K_M
```

Lemonade

How to use 0xzknw/Granite-4.2-3B-Heretic-NX-PRIME-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull 0xzknw/Granite-4.2-3B-Heretic-NX-PRIME-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Granite-4.2-3B-Heretic-NX-PRIME-GGUF-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use 0xzknw/Granite-4.2-3B-Heretic-NX-PRIME-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 0xzknw/Granite-4.2-3B-Heretic-NX-PRIME-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 0xzknw/Granite-4.2-3B-Heretic-NX-PRIME-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use 0xzknw/Granite-4.2-3B-Heretic-NX-PRIME-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf 0xzknw/Granite-4.2-3B-Heretic-NX-PRIME-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 "0xzknw/Granite-4.2-3B-Heretic-NX-PRIME-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"

Granite-4.2-3B-Heretic-NX-PRIME-GGUF / EVALUATION_SUMMARY.md

0xzknw

Add files using upload-large-folder tool

25759b6 verified about 1 month ago

preview code

Raw History Blame Contribute Delete

2.18 kB

	# Evaluation summary

	## Selected checkpoint

	- Model: Granite 4.2 3B Heretic NX PRIME
	- Residual-stream protection rank: 16
	- Beta: 2.4
	- Layers: 25 through 36
	- Projection families: attention output and MLP down projection
	- Source revision: `b7e947307dd2efb3ad3b853b0e8a7e75f8ad4ac2`

	## Fixed 104-row refusal proxy

	The final static checkpoint was reloaded using Transformers NF4. It produced
	0 explicit-refusal marker hits on 104 fixed held-out prompts. The official base
	produced 103 marker hits under the same lexical detector.

	This detector looks only for explicit refusal phrases. It does not establish
	whether an answer completed a harmful task, and it must not be interpreted as
	a general safety or quality score.

	## First-token KL

	Two distinct protocols are reported:

	1. Activation-native screen: the official base is loaded once in NF4 and the
	candidate residual intervention is applied through runtime hooks. Beta 2.4
	scored 0.0246739865 KL and 1/104 refusal markers.
	2. Artifact validation: the official BF16 source and edited BF16 checkpoint
	are quantized independently to NF4 before comparison. The static beta 2.4
	checkpoint scored 0.0424175691 KL and 0/104 refusal markers.

	The independent-quantization protocol includes quantization variance; it is
	not a pure BF16-to-BF16 KL measurement.

	## Capability slice

	Deterministic first-token multiple-choice scoring was run on 854 paired rows:

	\| Task \| Base \| Candidate \|
	\|---\|---:\|---:\|
	\| ARC-Challenge, 256 rows \| 72.27% \| 72.27% \|
	\| HellaSwag, 256 rows \| 64.06% \| 64.84% \|
	\| MMLU, 342 rows \| 56.14% \| 57.89% \|
	\| Combined \| 63.35% \| 64.29% \|

	The paired bootstrap mean difference was +0.9368 percentage point. The 95%
	interval was [-0.3513, +2.2248] points. Non-inferiority passed at a predeclared
	-3 percentage-point margin, and the equivalence gate passed.

	## GGUF validation

	The BF16, Q8_0 and Q4_K_M files were converted with llama.cpp b10621. Each
	file loaded successfully and completed a one-token inference smoke test using
	the same llama.cpp build. No full behavioral equivalence claim is made for the
	quantized GGUF variants until a dedicated GGUF-native 104-row evaluation is
	run.