Instructions to use NbAiLab/borealis2-26b-a4b-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 NbAiLab/borealis2-26b-a4b-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 NbAiLab/borealis2-26b-a4b-preview-gguf:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf NbAiLab/borealis2-26b-a4b-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 NbAiLab/borealis2-26b-a4b-preview-gguf:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf NbAiLab/borealis2-26b-a4b-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 NbAiLab/borealis2-26b-a4b-preview-gguf:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf NbAiLab/borealis2-26b-a4b-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 NbAiLab/borealis2-26b-a4b-preview-gguf:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf NbAiLab/borealis2-26b-a4b-preview-gguf:Q4_K_M

Use Docker

docker model run hf.co/NbAiLab/borealis2-26b-a4b-preview-gguf:Q4_K_M

LM Studio
Jan
Ollama
How to use NbAiLab/borealis2-26b-a4b-preview-gguf with Ollama:
```
ollama run hf.co/NbAiLab/borealis2-26b-a4b-preview-gguf:Q4_K_M
```
Unsloth Desktop

How to use NbAiLab/borealis2-26b-a4b-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 NbAiLab/borealis2-26b-a4b-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": "NbAiLab/borealis2-26b-a4b-preview-gguf:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use NbAiLab/borealis2-26b-a4b-preview-gguf with Docker Model Runner:
```
docker model run hf.co/NbAiLab/borealis2-26b-a4b-preview-gguf:Q4_K_M
```

Lemonade

How to use NbAiLab/borealis2-26b-a4b-preview-gguf with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull NbAiLab/borealis2-26b-a4b-preview-gguf:Q4_K_M

Run and chat with the model

lemonade run user.borealis2-26b-a4b-preview-gguf-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use NbAiLab/borealis2-26b-a4b-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 NbAiLab/borealis2-26b-a4b-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 NbAiLab/borealis2-26b-a4b-preview-gguf:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use NbAiLab/borealis2-26b-a4b-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 NbAiLab/borealis2-26b-a4b-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 "NbAiLab/borealis2-26b-a4b-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"

Borealis 2 Preview — Q8_0 GGUF

Q8_0 llama.cpp conversion of Borealis 2 Preview, including a matching Q8_0 Gemma 4 multimodal projector. Use borealis-2-preview-q8_0.gguf with mmproj-borealis-2-preview-q8_0.gguf in llama.cpp.

Lineage

This release is quantized directly from NbAiLab/nb-gpt-gemma4-26b-a4b-instruct-aurora-2604-pre-aurora-sft-2606-post-epoch2, the selected epoch-2 SFT checkpoint.

That checkpoint is based on google/gemma-4-26B-A4B-it, via the technical Aurora continued-pretraining and SFT lineage described in its source model card. Borealis 2 Preview is the public model name; “Aurora” is retained only when referring to that technical training lineage.

Both the language model and the multimodal projector are converted from the same epoch2 package. The included processor configuration is Gemma 4 compatible, preserving image input support in llama.cpp.

Files

borealis-2-preview-q8_0.gguf — Q8_0 language model
mmproj-borealis-2-preview-q8_0.gguf — matching Q8_0 Gemma 4 vision projector

Downloads last month: 162

GGUF

Model size

25B params

Architecture

gemma4

Hardware compatibility

3-bit

4-bit

5-bit

6-bit

8-bit

16-bit

Inference Providers NEW

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

Model tree for NbAiLab/borealis2-26b-a4b-preview-gguf

Base model

google/gemma-4-26B-A4B

Finetuned

google/gemma-4-26B-A4B-it

Finetuned

NbAiLab/borealis2-26b-a4b-preview

Quantized

(6)

this model

Collection including NbAiLab/borealis2-26b-a4b-preview-gguf

🌌 Borealis 2 Preview

Collection

Preview release of the Borealis 2 family of models by the National Library of Norway pre-trained in licensed material. • 2 items • Updated 1 day ago