Instructions to use dmvevents/nora-4b-v3.2-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 dmvevents/nora-4b-v3.2-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 dmvevents/nora-4b-v3.2-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf dmvevents/nora-4b-v3.2-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 dmvevents/nora-4b-v3.2-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf dmvevents/nora-4b-v3.2-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 dmvevents/nora-4b-v3.2-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf dmvevents/nora-4b-v3.2-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 dmvevents/nora-4b-v3.2-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf dmvevents/nora-4b-v3.2-GGUF:Q4_K_M

Use Docker

docker model run hf.co/dmvevents/nora-4b-v3.2-GGUF:Q4_K_M

LM Studio
Jan
Ollama
How to use dmvevents/nora-4b-v3.2-GGUF with Ollama:
```
ollama run hf.co/dmvevents/nora-4b-v3.2-GGUF:Q4_K_M
```
Unsloth Desktop

How to use dmvevents/nora-4b-v3.2-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf dmvevents/nora-4b-v3.2-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": "dmvevents/nora-4b-v3.2-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use dmvevents/nora-4b-v3.2-GGUF with Docker Model Runner:
```
docker model run hf.co/dmvevents/nora-4b-v3.2-GGUF:Q4_K_M
```

Lemonade

How to use dmvevents/nora-4b-v3.2-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull dmvevents/nora-4b-v3.2-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.nora-4b-v3.2-GGUF-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use dmvevents/nora-4b-v3.2-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 dmvevents/nora-4b-v3.2-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 dmvevents/nora-4b-v3.2-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use dmvevents/nora-4b-v3.2-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf dmvevents/nora-4b-v3.2-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 "dmvevents/nora-4b-v3.2-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"

You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

Nora v3.2 GGUF Quantizations

Laptop-deployable GGUF versions of dmvevents/nora-4b-v3.2.

File	Quant	Size	Use
`nora-v3.2-Q4_K_M.gguf`	Q4_K_M	2.9 GB	Tightest RAM budgets
`nora-v3.2-Q6_K.gguf`	Q6_K	3.8 GB	Recommended on 16 GB laptops — ~half the quantization loss of Q4_K_M (quant damage concentrates in Creole + math), 4.0 GB peak RAM / 10 tok/s measured at 6 CPU threads
`nora-v3.2-Q8_0.gguf`	Q8_0	4.9 GB	Near-lossless if RAM allows
`nora-v3.2-f16.gguf`	F16	9.1 GB	Reference / further quantization

Run with repeat_penalty=1.0 (the production default): higher values pressure paraphrase of verbatim facts (phone numbers, fees).

Eval (underlying bf16 model)

1,420-paraphrase eval: 89.2% Claude Sonnet 4.5 judge (v3.1 was 87.1%, v3 was 86.4%, v2 was 84.9%)
143-base eval: 89.9% Claude
Targeted gains vs v3.1: safety +5.03pp, gov +3.62pp, creole +2.60pp
See dmvevents/tt-eval-v3.2-results for full data.

Downloads last month: 1

GGUF

Model size

5B params

Architecture

qwen35

Hardware compatibility

4-bit

6-bit

8-bit

Inference Providers NEW

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

Model tree for dmvevents/nora-4b-v3.2-GGUF

Base model

Qwen/Qwen3.5-4B-Base

Finetuned

Qwen/Qwen3.5-4B

Finetuned

dmvevents/nora-4b-v3.2

Quantized

(1)

this model