Instructions to use QinEmPeRoR93/nassila-sanad-12b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use QinEmPeRoR93/nassila-sanad-12b with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="QinEmPeRoR93/nassila-sanad-12b",
	filename="nassila-sanad-12b-q6_k.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use QinEmPeRoR93/nassila-sanad-12b 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 QinEmPeRoR93/nassila-sanad-12b:Q6_K
# Run inference directly in the terminal:
llama cli -hf QinEmPeRoR93/nassila-sanad-12b:Q6_K

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf QinEmPeRoR93/nassila-sanad-12b:Q6_K
# Run inference directly in the terminal:
llama cli -hf QinEmPeRoR93/nassila-sanad-12b:Q6_K

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 QinEmPeRoR93/nassila-sanad-12b:Q6_K
# Run inference directly in the terminal:
./llama-cli -hf QinEmPeRoR93/nassila-sanad-12b:Q6_K

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 QinEmPeRoR93/nassila-sanad-12b:Q6_K
# Run inference directly in the terminal:
./build/bin/llama-cli -hf QinEmPeRoR93/nassila-sanad-12b:Q6_K

Use Docker

docker model run hf.co/QinEmPeRoR93/nassila-sanad-12b:Q6_K

LM Studio
Jan

vLLM

How to use QinEmPeRoR93/nassila-sanad-12b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "QinEmPeRoR93/nassila-sanad-12b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "QinEmPeRoR93/nassila-sanad-12b",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/QinEmPeRoR93/nassila-sanad-12b:Q6_K

Ollama
How to use QinEmPeRoR93/nassila-sanad-12b with Ollama:
```
ollama run hf.co/QinEmPeRoR93/nassila-sanad-12b:Q6_K
```

Unsloth Studio

How to use QinEmPeRoR93/nassila-sanad-12b 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 QinEmPeRoR93/nassila-sanad-12b 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 QinEmPeRoR93/nassila-sanad-12b to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for QinEmPeRoR93/nassila-sanad-12b to start chatting

How to use QinEmPeRoR93/nassila-sanad-12b with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf QinEmPeRoR93/nassila-sanad-12b:Q6_K

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": "QinEmPeRoR93/nassila-sanad-12b:Q6_K"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use QinEmPeRoR93/nassila-sanad-12b with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf QinEmPeRoR93/nassila-sanad-12b:Q6_K

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 QinEmPeRoR93/nassila-sanad-12b:Q6_K

Run Hermes

hermes

Atomic Chat new

OpenClaw new

How to use QinEmPeRoR93/nassila-sanad-12b with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf QinEmPeRoR93/nassila-sanad-12b:Q6_K

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 "QinEmPeRoR93/nassila-sanad-12b:Q6_K" \
  --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 QinEmPeRoR93/nassila-sanad-12b with Docker Model Runner:
```
docker model run hf.co/QinEmPeRoR93/nassila-sanad-12b:Q6_K
```

Lemonade

How to use QinEmPeRoR93/nassila-sanad-12b with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull QinEmPeRoR93/nassila-sanad-12b:Q6_K

Run and chat with the model

lemonade run user.nassila-sanad-12b-Q6_K

List all available models

lemonade list

Nassila Sanad 12B (quality tier)

Checkpoint: S14 (legacy train label v1.14)

Local GGUF for Sanad in Nassila — checks manuscript claims against source excerpts and returns structured JSON with verdicts and verbatim quotes.

File: nassila-sanad-12b-q6_k.gguf · Q6_K · ~9.1 GB
Default tier: nassila-sanad-e4b

Part of Nassila Ouroboros — see the E4B model card for the seven-worker overview.

Sanad today: validated on abstract excerpts (Tier 2). Full paper body text is planned (Tier 3).

Combined	Quote validity	False-supported
90.43%	100%	2.86%

Quality-tier validation: PASS

Usage

Quick start (Nassila + LM Studio)

Recommended — download this GGUF, load it in LM Studio, and start the Local Server at http://localhost:1234. ~12 GB+ VRAM recommended.

In Nassila: Settings → Passage grounding → runner LM Studio → model nassila-sanad-12b (or the id LM Studio shows).

Ollama

Requires Ollama 0.5+ and a public Hugging Face repo.

Pull from Hub:

ollama pull huggingface.co/QinEmPeRoR93/nassila-sanad-12b:Q6_K

In Nassila: runner Ollama → base URL http://localhost:11434 → model name from ollama list (often nassila-sanad-12b:Q6_K).

Modelfile fallback (private repo or pull tag not indexed)

FROM https://huggingface.co/QinEmPeRoR93/nassila-sanad-12b/resolve/main/nassila-sanad-12b-q6_k.gguf
PARAMETER num_ctx 4096

ollama create nassila-sanad-12b -f Modelfile

Advanced (llama.cpp / vLLM)

Serve the GGUF with any OpenAI-compatible server (ctx-size 4096; requires a recent llama.cpp build with gemma4_unified support). Point Nassila at your base URL and exposed model id.

llama-server -m nassila-sanad-12b-q6_k.gguf \
  --host 127.0.0.1 --port 1234 --ctx-size 4096 --n-gpu-layers 99

Limitations

Trained on abstract excerpts (Tier 2); full paper body (Tier 3) planned.
Advisory only — use with Nassila deterministic guardrails.
Not bundled in the Nassila installer.

Base model

google/gemma-4-12B-it · Gemma Terms of Use

Downloads last month: 37

GGUF

Model size

12B params

Architecture

gemma4

Hardware compatibility

6-bit

Model tree for QinEmPeRoR93/nassila-sanad-12b

Base model

google/gemma-4-12B

Finetuned

google/gemma-4-12B-it

Quantized

(267)

this model