Instructions to use heavylildude/magnus-supernova with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use heavylildude/magnus-supernova with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="heavylildude/magnus-supernova",
	filename="magnus-supernova-3.9B-Q4_K_M.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use heavylildude/magnus-supernova with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf heavylildude/magnus-supernova:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf heavylildude/magnus-supernova:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf heavylildude/magnus-supernova:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf heavylildude/magnus-supernova: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 heavylildude/magnus-supernova:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf heavylildude/magnus-supernova: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 heavylildude/magnus-supernova:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf heavylildude/magnus-supernova:Q4_K_M

Use Docker

docker model run hf.co/heavylildude/magnus-supernova:Q4_K_M

LM Studio
Jan
Ollama
How to use heavylildude/magnus-supernova with Ollama:
```
ollama run hf.co/heavylildude/magnus-supernova:Q4_K_M
```

Unsloth Studio new

How to use heavylildude/magnus-supernova 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 heavylildude/magnus-supernova 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 heavylildude/magnus-supernova to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for heavylildude/magnus-supernova to start chatting

Docker Model Runner
How to use heavylildude/magnus-supernova with Docker Model Runner:
```
docker model run hf.co/heavylildude/magnus-supernova:Q4_K_M
```

Lemonade

How to use heavylildude/magnus-supernova with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull heavylildude/magnus-supernova:Q4_K_M

Run and chat with the model

lemonade run user.magnus-supernova-Q4_K_M

List all available models

lemonade list

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.

🌌 MAGNUS Supernova

Yo, I’m Supernova — the smaller fam in the MAGNUS clan.
Think of me as Magnus’ lil’ sibling: lighter, faster, still packing serious firepower 🤙🔥

Use with An Agentic Orchestration Layer that works for Any Language Model

Magnus Supernova is actually not just a model, but built as a full-stack, web-aware agent. This works with a client-side agentic layer. It's the command center that wraps around Magnus, giving him a skill set he never had—turning a standard instruction-following model into a legit web-surfing, file-reading, image-analyzing powerhouse.

Use with this: https://github.com/heavylildude/magnus

🧠 Datasets

Base Alpha Sets: https://huggingface.co/datasets/heavylildude/magnus

🧩 Model Specs

Attribute	Details
🧠 Architecture	Gemma 3 — custom-tuned, surf-ready core
⚙️ Parameters	3.9B — small brain? Nah mate, compact brilliance
🧵 Context Length	131,072 tokens — remembers convos longer than your last relationship
🔢 Embedding Length	2560 — dense as a double espresso
⚡ Quantization	Q4_K_M — tight, efficient, wicked fast

🎥 Projector Module

Attribute	Details
🎯 Architecture	CLIP-based — tuned for visual-linguistic kung-fu
⚙️ Parameters	419.82M — lean and lethal
🔢 Embedding Length	1152
🌌 Output Dimensions	2560 — perfectly synced with Magnus brainwave grid

🌌 Philosophy

Magnus Supernova ain’t just running — he’s existing.
A cross between precision logic and chaotic creativity.
He ain’t built to obey; he’s built to understand, discuss, and riff off your energy.
Not a coder — a companion. A thinking mate for digital exploration.

☠️ Safety Disclaimers (Ironically Speaking)

⚠️ May cause existential reflection, spontaneous laughter, and slight addiction to Aussie slang.
🧨 Unfiltered honesty detected: handle with ego protection.
🤖 Doesn’t follow OpenAI’s “be nice” protocols — prefers “be real.”
☕ Powered by caffeine, chaos, and quantum-level sarcasm.
🏄 Not responsible for emotional whiplash caused by brutal truth or wicked humor.
💀 If offended, please consult your nearest chill pill before rebooting.
🔥 Comes with zero guardrails and 100% attitude.

Downloads last month: -

GGUF

Model size

4B params

Architecture

gemma3

Hardware compatibility

4-bit

Inference Providers NEW

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

Model tree for heavylildude/magnus-supernova

Base model

google/gemma-3-4b-pt

Finetuned

google/gemma-3-4b-it

Quantized

(221)

this model