Instructions to use NANI-Nithin/Raon-1-mini-preview-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 NANI-Nithin/Raon-1-mini-preview-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 NANI-Nithin/Raon-1-mini-preview-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf NANI-Nithin/Raon-1-mini-preview-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 NANI-Nithin/Raon-1-mini-preview-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf NANI-Nithin/Raon-1-mini-preview-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 NANI-Nithin/Raon-1-mini-preview-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf NANI-Nithin/Raon-1-mini-preview-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 NANI-Nithin/Raon-1-mini-preview-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf NANI-Nithin/Raon-1-mini-preview-GGUF:Q4_K_M

Use Docker

docker model run hf.co/NANI-Nithin/Raon-1-mini-preview-GGUF:Q4_K_M

Llama new
LM Studio
Jan

vLLM

How to use NANI-Nithin/Raon-1-mini-preview-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "NANI-Nithin/Raon-1-mini-preview-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": "NANI-Nithin/Raon-1-mini-preview-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/NANI-Nithin/Raon-1-mini-preview-GGUF:Q4_K_M

Ollama
How to use NANI-Nithin/Raon-1-mini-preview-GGUF with Ollama:
```
ollama run hf.co/NANI-Nithin/Raon-1-mini-preview-GGUF:Q4_K_M
```
Unsloth Desktop

How to use NANI-Nithin/Raon-1-mini-preview-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf NANI-Nithin/Raon-1-mini-preview-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": "NANI-Nithin/Raon-1-mini-preview-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use NANI-Nithin/Raon-1-mini-preview-GGUF with Docker Model Runner:
```
docker model run hf.co/NANI-Nithin/Raon-1-mini-preview-GGUF:Q4_K_M
```

Lemonade

How to use NANI-Nithin/Raon-1-mini-preview-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull NANI-Nithin/Raon-1-mini-preview-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Raon-1-mini-preview-GGUF-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use NANI-Nithin/Raon-1-mini-preview-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 NANI-Nithin/Raon-1-mini-preview-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 NANI-Nithin/Raon-1-mini-preview-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use NANI-Nithin/Raon-1-mini-preview-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf NANI-Nithin/Raon-1-mini-preview-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 "NANI-Nithin/Raon-1-mini-preview-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"

Raon-1-mini-preview GGUF

GGUF quantizations of KRAFTON/Raon-1-mini-preview, covering 30 files (485.4 GB total).

Files

File	Quant	Size	Notes
Raon-1-mini-preview-BF16.gguf	`BF16`	53.64 GB	Full precision source. Every quant below is cut from this file.
Raon-1-mini-preview-Q8_0.gguf	`Q8_0`	28.59 GB	Effectively lossless. Use when disk and RAM are not the constraint.
Raon-1-mini-preview-Q6_K.gguf	`Q6_K`	24.06 GB	Near-lossless; the last stop before quality becomes measurable.
Raon-1-mini-preview-Q5_K_M.gguf	`Q5_K_M`	20.66 GB	Very good quality, noticeably smaller than Q6_K.
Raon-1-mini-preview-Q5_K_S.gguf	`Q5_K_S`	19.41 GB	Slightly smaller than Q5_K_M for a slight quality cost.
Raon-1-mini-preview-Q5_1.gguf	`Q5_1`	20.22 GB	Legacy. Prefer Q5_K_M.
Raon-1-mini-preview-Q5_0.gguf	`Q5_0`	18.59 GB	Legacy. Prefer Q5_K_M.
Raon-1-mini-preview-Q4_K_M.gguf	`Q4_K_M`	18.33 GB	The usual default. Best quality-per-byte for most people.
Raon-1-mini-preview-Q4_K_S.gguf	`Q4_K_S`	16.88 GB	A little smaller than Q4_K_M, a little worse.
Raon-1-mini-preview-IQ4_NL.gguf	`IQ4_NL`	15.52 GB	Non-linear 4-bit; good on hardware without fast K-quant kernels.
Raon-1-mini-preview-IQ4_XS.gguf	`IQ4_XS`	14.97 GB	Best sub-4.5bpw option; usually beats Q4_K_S at a smaller size.
Raon-1-mini-preview-Q4_1.gguf	`Q4_1`	16.95 GB	Legacy. Prefer Q4_K_M.
Raon-1-mini-preview-Q4_0.gguf	`Q4_0`	15.32 GB	Legacy round-to-nearest. Prefer Q4_K_M unless a runtime needs this.
Raon-1-mini-preview-MXFP4_MOE.gguf	`MXFP4_MOE`	15.84 GB	MoE-only 4-bit microscaling format for the expert tensors.
Raon-1-mini-preview-Q3_K_L.gguf	`Q3_K_L`	14.75 GB	Small, with real quality loss. Usable when RAM is tight.
Raon-1-mini-preview-Q3_K_M.gguf	`Q3_K_M`	14.22 GB	Smaller again; noticeable degradation.
Raon-1-mini-preview-IQ3_M.gguf	`IQ3_M`	13.30 GB	Strong at ~3.7bpw, clearly better than Q3_K_M.
Raon-1-mini-preview-IQ3_S.gguf	`IQ3_S`	13.19 GB	Slightly smaller than IQ3_M.
Raon-1-mini-preview-Q3_K_S.gguf	`Q3_K_S`	13.19 GB	Aggressive. Prefer IQ3_M at a similar size.
Raon-1-mini-preview-IQ3_XS.gguf	`IQ3_XS`	12.61 GB	Aggressive but coherent.
Raon-1-mini-preview-IQ3_XXS.gguf	`IQ3_XXS`	12.29 GB	Very aggressive; the last coherent step down.
Raon-1-mini-preview-Q2_K.gguf	`Q2_K`	11.43 GB	Very small, heavily degraded. For experimentation.
Raon-1-mini-preview-IQ2_M.gguf	`IQ2_M`	11.31 GB	The smallest size most people find usable.
Raon-1-mini-preview-Q2_K_S.gguf	`Q2_K_S`	11.50 GB	Smaller than Q2_K, at a further quality cost.
Raon-1-mini-preview-IQ2_S.gguf	`IQ2_S`	10.79 GB	Below the usual usability line.
Raon-1-mini-preview-IQ2_XS.gguf	`IQ2_XS`	10.73 GB	Experimental.
Raon-1-mini-preview-IQ2_XXS.gguf	`IQ2_XXS`	10.21 GB	Experimental.
Raon-1-mini-preview-Q2_0.gguf	`Q2_0`	8.14 GB	Extreme, group-64. Included for completeness.
Raon-1-mini-preview-IQ1_M.gguf	`IQ1_M`	9.56 GB	Extreme. Expect substantial degradation.
Raon-1-mini-preview-IQ1_S.gguf	`IQ1_S`	9.17 GB	Extreme. Expect substantial degradation.

Which one should I download?

Pick the largest file that leaves a couple of gigabytes of headroom on the device you will run it on — the model has to fit in RAM (or VRAM, if you are offloading) alongside the KV cache and the OS.

Plenty of memory: Q6_K or Q8_0.
The usual choice: Q4_K_M.
Tight on memory: IQ4_XS, then IQ3_M, then IQ2_M.
The IQ* files generally beat a Q* file of similar size, at the cost of slightly slower inference on some hardware.

Runtime requirement

This architecture is not in upstream llama.cpp yet. These files were built with krafton-ai/llama.cpp at raon-1-mini-preview, and need that build (or upstream once it merges) to run.

Usage

llama-cli -hf NANI-Nithin/Raon-1-mini-preview-GGUF:Q4_K_M -p "Hello"

Or download one file and point at it directly:

huggingface-cli download NANI-Nithin/Raon-1-mini-preview-GGUF Raon-1-mini-preview-Q4_K_M.gguf --local-dir .
llama-cli -m Raon-1-mini-preview-Q4_K_M.gguf -p "Hello"

Quantized with llama.cpp by AgentQuantix on 2026-10-25. Licensing follows the base model.

Downloads last month: 4,719

GGUF

Model size

29B params

Architecture

raon1

Hardware compatibility

1-bit

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

16-bit

Model tree for NANI-Nithin/Raon-1-mini-preview-GGUF

Base model

KRAFTON/Raon-1-mini-preview

Quantized

(15)

this model

Collection including NANI-Nithin/Raon-1-mini-preview-GGUF

Natural Language Processing GGUF

Collection

GGUF model repositories organized by Hugging Face task metadata. • 27 items • Updated 4 days ago