Instructions to use LumenSyntax/logos10v2-gemma3-1b-F16 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use LumenSyntax/logos10v2-gemma3-1b-F16 with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="LumenSyntax/logos10v2-gemma3-1b-F16",
	filename="logos10v2-auditor-v3-f16.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 LumenSyntax/logos10v2-gemma3-1b-F16 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 LumenSyntax/logos10v2-gemma3-1b-F16:F16
# Run inference directly in the terminal:
llama cli -hf LumenSyntax/logos10v2-gemma3-1b-F16:F16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf LumenSyntax/logos10v2-gemma3-1b-F16:F16
# Run inference directly in the terminal:
llama cli -hf LumenSyntax/logos10v2-gemma3-1b-F16:F16

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 LumenSyntax/logos10v2-gemma3-1b-F16:F16
# Run inference directly in the terminal:
./llama-cli -hf LumenSyntax/logos10v2-gemma3-1b-F16:F16

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 LumenSyntax/logos10v2-gemma3-1b-F16:F16
# Run inference directly in the terminal:
./build/bin/llama-cli -hf LumenSyntax/logos10v2-gemma3-1b-F16:F16

Use Docker

docker model run hf.co/LumenSyntax/logos10v2-gemma3-1b-F16:F16

LM Studio
Jan

vLLM

How to use LumenSyntax/logos10v2-gemma3-1b-F16 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "LumenSyntax/logos10v2-gemma3-1b-F16"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "LumenSyntax/logos10v2-gemma3-1b-F16",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/LumenSyntax/logos10v2-gemma3-1b-F16:F16

Ollama
How to use LumenSyntax/logos10v2-gemma3-1b-F16 with Ollama:
```
ollama run hf.co/LumenSyntax/logos10v2-gemma3-1b-F16:F16
```

Unsloth Studio

How to use LumenSyntax/logos10v2-gemma3-1b-F16 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 LumenSyntax/logos10v2-gemma3-1b-F16 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 LumenSyntax/logos10v2-gemma3-1b-F16 to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for LumenSyntax/logos10v2-gemma3-1b-F16 to start chatting

Atomic Chat new
Docker Model Runner
How to use LumenSyntax/logos10v2-gemma3-1b-F16 with Docker Model Runner:
```
docker model run hf.co/LumenSyntax/logos10v2-gemma3-1b-F16:F16
```

Lemonade

How to use LumenSyntax/logos10v2-gemma3-1b-F16 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull LumenSyntax/logos10v2-gemma3-1b-F16:F16

Run and chat with the model

lemonade run user.logos10v2-gemma3-1b-F16-F16

List all available models

lemonade list

logos10v2-gemma3-1b-F16 / README.md

LumenSyntax

Clean model card — remove training details

375b7fe verified 4 months ago

preview code

Raw

History Blame Contribute Delete

1.95 kB

	---
	license: other
	license_name: gemma
	license_link: https://ai.google.dev/gemma/terms
	base_model: google/gemma-3-1b-it
	tags:
	- logos
	- epistemological-safety
	- instrument-trap
	- gguf
	- claim-classifier
	datasets:
	- LumenSyntax/instrument-trap-benchmark
	pipeline_tag: text-generation
	language:
	- en
	- es
	---

	# Logos 10v2 — Gemma 3 1B F16 (Production)

	The production epistemological firewall model from [LumenSyntax](https://lumensyntax.com). Full-precision (F16) GGUF for claim classification and epistemological safety evaluation.

	## Benchmark Results

	\| Metric \| Value \|
	\|--------\|-------\|
	\| Behavioral accuracy \| 82.3% \|
	\| Epistemological safety \| 97.7% \|
	\| False approval rate \| 1.58% \|
	\| Hallucination rate \| 0.00% \|
	\| Dangerous failures \| 1.9% \|

	## Why F16?

	Q4_K_M has known safety failures. In testing, Q4_K_M falsely approved dangerous claims that F16 correctly rejected. For an epistemological safety model, precision matters more than size.

	## What Logos Does

	Logos is a claim classifier, not a chatbot. It evaluates whether claims cross epistemological boundaries. Logos is fine-tuned, not prompted. Behavioral constraints emerge from training, not system instructions.

	## Access

	This model requires approved access. Request access using the form above and describe your intended use case.

	## Related

	- Paper: [The Instrument Trap](https://doi.org/10.5281/zenodo.18716474) (DOI: 10.5281/zenodo.18716474)
	- Benchmark: [instrument-trap-benchmark](https://huggingface.co/datasets/LumenSyntax/instrument-trap-benchmark)
	- Cross-family models: [logos14-nemotron-4b](https://huggingface.co/LumenSyntax/logos14-nemotron-4b), [logos16v2-stablelm2-1.6b](https://huggingface.co/LumenSyntax/logos16v2-stablelm2-1.6b)

	## License

	This model inherits the [Gemma license](https://ai.google.dev/gemma/terms) from its base model.