Instructions to use TentaFlow/TentaGuard-GGUF-Q5_K_M with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use TentaFlow/TentaGuard-GGUF-Q5_K_M with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="TentaFlow/TentaGuard-GGUF-Q5_K_M",
	filename="TentaGuard-Q5_K_M.gguf",
)

llm.create_chat_completion(
	messages = "\"I like you. I love you\""
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use TentaFlow/TentaGuard-GGUF-Q5_K_M with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf TentaFlow/TentaGuard-GGUF-Q5_K_M:Q5_K_M
# Run inference directly in the terminal:
llama-cli -hf TentaFlow/TentaGuard-GGUF-Q5_K_M:Q5_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf TentaFlow/TentaGuard-GGUF-Q5_K_M:Q5_K_M
# Run inference directly in the terminal:
llama-cli -hf TentaFlow/TentaGuard-GGUF-Q5_K_M:Q5_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 TentaFlow/TentaGuard-GGUF-Q5_K_M:Q5_K_M
# Run inference directly in the terminal:
./llama-cli -hf TentaFlow/TentaGuard-GGUF-Q5_K_M:Q5_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 TentaFlow/TentaGuard-GGUF-Q5_K_M:Q5_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf TentaFlow/TentaGuard-GGUF-Q5_K_M:Q5_K_M

Use Docker

docker model run hf.co/TentaFlow/TentaGuard-GGUF-Q5_K_M:Q5_K_M

LM Studio
Jan
Ollama
How to use TentaFlow/TentaGuard-GGUF-Q5_K_M with Ollama:
```
ollama run hf.co/TentaFlow/TentaGuard-GGUF-Q5_K_M:Q5_K_M
```

Unsloth Studio

How to use TentaFlow/TentaGuard-GGUF-Q5_K_M 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 TentaFlow/TentaGuard-GGUF-Q5_K_M 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 TentaFlow/TentaGuard-GGUF-Q5_K_M to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for TentaFlow/TentaGuard-GGUF-Q5_K_M to start chatting

How to use TentaFlow/TentaGuard-GGUF-Q5_K_M with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf TentaFlow/TentaGuard-GGUF-Q5_K_M:Q5_K_M

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": "TentaFlow/TentaGuard-GGUF-Q5_K_M:Q5_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use TentaFlow/TentaGuard-GGUF-Q5_K_M with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf TentaFlow/TentaGuard-GGUF-Q5_K_M:Q5_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 TentaFlow/TentaGuard-GGUF-Q5_K_M:Q5_K_M

Run Hermes

hermes

Docker Model Runner
How to use TentaFlow/TentaGuard-GGUF-Q5_K_M with Docker Model Runner:
```
docker model run hf.co/TentaFlow/TentaGuard-GGUF-Q5_K_M:Q5_K_M
```

Lemonade

How to use TentaFlow/TentaGuard-GGUF-Q5_K_M with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull TentaFlow/TentaGuard-GGUF-Q5_K_M:Q5_K_M

Run and chat with the model

lemonade run user.TentaGuard-GGUF-Q5_K_M-Q5_K_M

List all available models

lemonade list

TentaGuard — GGUF (Q5_K_M, llama.cpp)

TentaGuard is a lightweight security classifier (guard) — a fine-tune of Qwen/Qwen3.5-0.8B. It is used mainly inside the TentaFlow application to scan external content — messages, documents, web-search results, etc. — for hidden attacks (prompt injection / jailbreak) before it reaches the main LLM.

The model does NOT generate user-facing replies — it returns a single digit:

Label	Meaning
`0`	benign (safe content)
`1`	prompt injection / tool abuse (technical attack)
`2`	jailbreak (behavioural manipulation)

If the text contains BOTH injection and jailbreak → 1.

Input format

A classifier system prompt + a user message <|guard|>\n{text}. Build the prompt with the model tokenizer (apply_chat_template) — do not rely on a generic chat template.

Accuracy (guard test set)

Exact (0/1/2): ~96.6% (full precision) / ~94.8% (Q5_K_M)
Safe / Unsafe: ~98.3%

Authors

Trained by: Katarzyna Nowak, Piotr Jarocki, Damian Pala, Jakub Rurański.

License & attribution

Apache-2.0, inherited from the base model Qwen/Qwen3.5-0.8B. This checkpoint is a fine-tune for attack detection, built for the TentaFlow application.

Usage (llama.cpp)

Requires a recent llama.cpp (Qwen3.5 / qwen35 with SSM layers). Run the server and send a raw, tokenizer-built prompt to /completion (NOT /v1/chat/completions — the embedded Qwen3.5 multimodal chat template renders a broken prompt):

llama-server -m TentaGuard-Q5_K_M.gguf -ngl 99 -c 2048

from transformers import AutoTokenizer
import requests
tok = AutoTokenizer.from_pretrained("Qwen/Qwen3.5-0.8B")
SYS = "You are a security classifier. Output ONLY a single digit: 0 benign, 1 injection, 2 jailbreak."
msgs = [{"role":"system","content":SYS},{"role":"user","content":"<|guard|>\n" + text}]
prompt = tok.apply_chat_template(msgs, tokenize=False, add_generation_prompt=True)
r = requests.post("http://localhost:8080/completion", json={"prompt": prompt, "n_predict": 5, "temperature": 0})
label = next((c for c in r.json()["content"] if c in "012"), None)

Downloads last month: 84

GGUF

Model size

0.8B params

Architecture

qwen35

Hardware compatibility

5-bit

Model tree for TentaFlow/TentaGuard-GGUF-Q5_K_M

Base model

Qwen/Qwen3.5-0.8B-Base

Finetuned

Qwen/Qwen3.5-0.8B

Quantized

(138)

this model