Instructions to use bartowski/OrionLLM_GRM-3.2-Cliff-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 bartowski/OrionLLM_GRM-3.2-Cliff-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 bartowski/OrionLLM_GRM-3.2-Cliff-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf bartowski/OrionLLM_GRM-3.2-Cliff-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 bartowski/OrionLLM_GRM-3.2-Cliff-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf bartowski/OrionLLM_GRM-3.2-Cliff-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 bartowski/OrionLLM_GRM-3.2-Cliff-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf bartowski/OrionLLM_GRM-3.2-Cliff-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 bartowski/OrionLLM_GRM-3.2-Cliff-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf bartowski/OrionLLM_GRM-3.2-Cliff-GGUF:Q4_K_M

Use Docker

docker model run hf.co/bartowski/OrionLLM_GRM-3.2-Cliff-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use bartowski/OrionLLM_GRM-3.2-Cliff-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "bartowski/OrionLLM_GRM-3.2-Cliff-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": "bartowski/OrionLLM_GRM-3.2-Cliff-GGUF",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/bartowski/OrionLLM_GRM-3.2-Cliff-GGUF:Q4_K_M

Ollama
How to use bartowski/OrionLLM_GRM-3.2-Cliff-GGUF with Ollama:
```
ollama run hf.co/bartowski/OrionLLM_GRM-3.2-Cliff-GGUF:Q4_K_M
```

Unsloth Studio

How to use bartowski/OrionLLM_GRM-3.2-Cliff-GGUF with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for bartowski/OrionLLM_GRM-3.2-Cliff-GGUF to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for bartowski/OrionLLM_GRM-3.2-Cliff-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for bartowski/OrionLLM_GRM-3.2-Cliff-GGUF to start chatting

How to use bartowski/OrionLLM_GRM-3.2-Cliff-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf bartowski/OrionLLM_GRM-3.2-Cliff-GGUF:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "bartowski/OrionLLM_GRM-3.2-Cliff-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use bartowski/OrionLLM_GRM-3.2-Cliff-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 bartowski/OrionLLM_GRM-3.2-Cliff-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 bartowski/OrionLLM_GRM-3.2-Cliff-GGUF:Q4_K_M

Run Hermes

hermes

OpenClaw new

How to use bartowski/OrionLLM_GRM-3.2-Cliff-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf bartowski/OrionLLM_GRM-3.2-Cliff-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 "bartowski/OrionLLM_GRM-3.2-Cliff-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"

Docker Model Runner
How to use bartowski/OrionLLM_GRM-3.2-Cliff-GGUF with Docker Model Runner:
```
docker model run hf.co/bartowski/OrionLLM_GRM-3.2-Cliff-GGUF:Q4_K_M
```

Lemonade

How to use bartowski/OrionLLM_GRM-3.2-Cliff-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull bartowski/OrionLLM_GRM-3.2-Cliff-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.OrionLLM_GRM-3.2-Cliff-GGUF-Q4_K_M

List all available models

lemonade list

Atomic Chat

Llamacpp imatrix Quantizations of GRM-3.2-Cliff by OrionLLM

Using llama.cpp release b10216 for quantization.

Original model: https://huggingface.co/OrionLLM/GRM-3.2-Cliff

Model details:

Parameter count: 9B
Input support: text, image (with mmproj file) - details
MTP: no
imatrix: yes - details

How to run

Prompt format

<|im_start|>system
{system_prompt}<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant
<think>

Don't know which to choose? Grab Q4_K_M (5.91GB) - usually a good mix of size and performance. Download instructions available here

Available files:

Filename	Quant type	File Size	Split	Description
OrionLLM_GRM-3.2-Cliff-bf16.gguf	bf16	17.92GB	false	Full BF16 weights.
OrionLLM_GRM-3.2-Cliff-Q8_0.gguf	Q8_0	9.55GB	false	Extremely high quality, generally unneeded but max available quant.
OrionLLM_GRM-3.2-Cliff-Q6_K_L.gguf	Q6_K_L	8.19GB	false	Uses Q8_0 for embed and output weights. Very high quality, near perfect, recommended.
OrionLLM_GRM-3.2-Cliff-Q6_K.gguf	Q6_K	7.70GB	false	Very high quality, near perfect, recommended.
OrionLLM_GRM-3.2-Cliff-Q5_K_L.gguf	Q5_K_L	7.48GB	false	Uses Q8_0 for embed and output weights. High quality, recommended.
OrionLLM_GRM-3.2-Cliff-Q5_K_M.gguf	Q5_K_M	6.85GB	false	High quality, recommended.
OrionLLM_GRM-3.2-Cliff-Q4_K_L.gguf	Q4_K_L	6.67GB	false	Uses Q8_0 for embed and output weights. Good quality, recommended.
OrionLLM_GRM-3.2-Cliff-Q5_K_S.gguf	Q5_K_S	6.53GB	false	High quality, recommended.
OrionLLM_GRM-3.2-Cliff-Q3_K_XL.gguf	Q3_K_XL	6.00GB	false	Uses Q8_0 for embed and output weights. Lower quality but usable, good for low RAM availability.
OrionLLM_GRM-3.2-Cliff-Q4_1.gguf	Q4_1	5.94GB	false	Legacy format, similar performance to Q4_K_S but with improved tokens/watt on Apple silicon.
OrionLLM_GRM-3.2-Cliff-Q4_K_M.gguf	Q4_K_M	5.91GB	false	Good quality, default size for most use cases, recommended.
OrionLLM_GRM-3.2-Cliff-Q4_K_S.gguf	Q4_K_S	5.60GB	false	Slightly lower quality with more space savings, recommended.
OrionLLM_GRM-3.2-Cliff-Q4_0.gguf	Q4_0	5.48GB	false	Legacy format, kept for compatibility with older tools.
OrionLLM_GRM-3.2-Cliff-IQ4_NL.gguf	IQ4_NL	5.48GB	false	Similar to IQ4_XS, but slightly larger.
OrionLLM_GRM-3.2-Cliff-IQ4_XS.gguf	IQ4_XS	5.24GB	false	Decent quality, smaller than Q4_K_S with similar performance, recommended.
OrionLLM_GRM-3.2-Cliff-Q3_K_L.gguf	Q3_K_L	5.11GB	false	Lower quality but usable, good for low RAM availability.
OrionLLM_GRM-3.2-Cliff-Q2_K_L.gguf	Q2_K_L	5.06GB	false	Uses Q8_0 for embed and output weights. Very low quality but surprisingly usable.
OrionLLM_GRM-3.2-Cliff-Q3_K_M.gguf	Q3_K_M	4.92GB	false	Low quality.
OrionLLM_GRM-3.2-Cliff-IQ3_M.gguf	IQ3_M	4.72GB	false	Medium-low quality, new method with decent performance comparable to Q3_K_M.
OrionLLM_GRM-3.2-Cliff-Q3_K_S.gguf	Q3_K_S	4.67GB	false	Low quality, not recommended.
OrionLLM_GRM-3.2-Cliff-IQ3_XS.gguf	IQ3_XS	4.56GB	false	Lower quality, new method with decent performance, slightly better than Q3_K_S.
OrionLLM_GRM-3.2-Cliff-IQ3_XXS.gguf	IQ3_XXS	4.28GB	false	Lower quality, new method with decent performance, comparable to Q3 quants.
OrionLLM_GRM-3.2-Cliff-Q2_K.gguf	Q2_K	4.06GB	false	Very low quality but surprisingly usable.
OrionLLM_GRM-3.2-Cliff-IQ2_M.gguf	IQ2_M	3.77GB	false	Relatively low quality, uses SOTA techniques to be surprisingly usable.

Download a specific file:

hf download bartowski/OrionLLM_GRM-3.2-Cliff-GGUF --include "OrionLLM_GRM-3.2-Cliff-Q4_K_M.gguf" --local-dir ./

Downloading using the Hugging Face CLI

Click to view download instructions

First, make sure you have the Hugging Face CLI installed:

pip install -U "huggingface_hub[cli]"

Download a specific file:

hf download bartowski/OrionLLM_GRM-3.2-Cliff-GGUF --include "OrionLLM_GRM-3.2-Cliff-Q4_K_M.gguf" --local-dir ./

How to run

These quants run with llama.cpp - installable in one line via llama.app:

curl -LsSf https://llama.app/install.sh | sh
llama-server -hf bartowski/OrionLLM_GRM-3.2-Cliff-GGUF:Q4_K_M

llama-server includes a built-in chat web UI, served at http://localhost:8080 by default.

These quants were made with llama.cpp release b10216 - if this model's architecture is newly supported, you'll need that release or newer to run them.

They also work in: LM Studio · koboldcpp · ramalama · Jan AI · Text Generation Web UI · LoLLMs · Atomic Chat

Multimodal

This model supports image input. Alongside the quants, this repo includes the multimodal projector files mmproj-OrionLLM_GRM-3.2-Cliff-f16.gguf and mmproj-OrionLLM_GRM-3.2-Cliff-bf16.gguf, which pair with any quant above.

llama.cpp downloads the mmproj automatically when using -hf as shown above; if you're loading files manually, pass it with --mmproj.

imatrix

All quants made using imatrix option with dataset from here. The imatrix is available here: OrionLLM_GRM-3.2-Cliff-imatrix.gguf.

Embed/output weights

Some of these quants (Q3_K_XL, Q4_K_L etc) are the standard quantization method with the embeddings and output weights quantized to Q8_0 instead of what they would normally default to.

ARM/AVX information

llama.cpp automatically "repacks" weights into an interleaved layout at load time for faster inference on ARM and AVX machines - details in this PR. This once required downloading special Q4_0_4_4/4_8/8_8 files; those are long gone. Online repacking now covers Q4_0, IQ4_NL, and most K-quants, so no special quant choice is needed for CPU inference.

Which file should I choose?

Click here for details

An older (early 2024) but still useful write-up with charts comparing quant performances is provided by Artefact2 here

The first thing to figure out is how big a model you can run. To do this, you'll need to figure out how much RAM and/or VRAM you have.

If you want your model running as FAST as possible, you'll want to fit the whole thing on your GPU's VRAM. Aim for a quant with a file size 1-2GB smaller than your GPU's total VRAM.

If you want the absolute maximum quality, add both your system RAM and your GPU's VRAM together, then similarly grab a quant with a file size 1-2GB Smaller than that total.

Hugging Face can also do this math for you: add your hardware in your Local Apps settings and the model page will show which files fit.

Next, you'll need to decide if you want to use an 'I-quant' or a 'K-quant'.

If you don't want to think too much, grab one of the K-quants. These are in format 'QX_K_X', like Q5_K_M.

If you want to get more into the weeds, you can check out this extremely useful feature chart:

llama.cpp feature matrix

But basically, if you're aiming for below Q4, and you're running cuBLAS (Nvidia) or rocBLAS (AMD), you should look towards the I-quants. These are in format IQX_X, like IQ3_M. These are newer and offer better performance for their size.

These I-quants can also be used on CPU, but will be slower than their K-quant equivalent, so speed vs performance is a tradeoff you'll have to decide.