Instructions to use hacnho/gguf-add-eos-token-poc with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use hacnho/gguf-add-eos-token-poc with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="hacnho/gguf-add-eos-token-poc",
	filename="llama-spm-add-eos.gguf",
)

output = llm(
	"Once upon a time,",
	max_tokens=512,
	echo=True
)
print(output)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use hacnho/gguf-add-eos-token-poc 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 hacnho/gguf-add-eos-token-poc
# Run inference directly in the terminal:
llama cli -hf hacnho/gguf-add-eos-token-poc

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf hacnho/gguf-add-eos-token-poc
# Run inference directly in the terminal:
llama cli -hf hacnho/gguf-add-eos-token-poc

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 hacnho/gguf-add-eos-token-poc
# Run inference directly in the terminal:
./llama-cli -hf hacnho/gguf-add-eos-token-poc

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 hacnho/gguf-add-eos-token-poc
# Run inference directly in the terminal:
./build/bin/llama-cli -hf hacnho/gguf-add-eos-token-poc

Use Docker

docker model run hf.co/hacnho/gguf-add-eos-token-poc

LM Studio
Jan
Ollama
How to use hacnho/gguf-add-eos-token-poc with Ollama:
```
ollama run hf.co/hacnho/gguf-add-eos-token-poc
```

Unsloth Studio

How to use hacnho/gguf-add-eos-token-poc 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 hacnho/gguf-add-eos-token-poc 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 hacnho/gguf-add-eos-token-poc to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for hacnho/gguf-add-eos-token-poc to start chatting

Atomic Chat new
Docker Model Runner
How to use hacnho/gguf-add-eos-token-poc with Docker Model Runner:
```
docker model run hf.co/hacnho/gguf-add-eos-token-poc
```

Lemonade

How to use hacnho/gguf-add-eos-token-poc with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull hacnho/gguf-add-eos-token-poc

Run and chat with the model

lemonade run user.gguf-add-eos-token-poc-{{QUANT_TAG}}

List all available models

lemonade list

YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

GGUF add_eos_token PoC

This repo is a benign security research PoC for Huntr MFV triage.

Files:

llama-spm-add-eos.gguf
verify_remote_poc.py

Tested trigger entrypoint:

llama-tokenize -m <model> --ids -p <prompt>

Behavior:

seed vocab-only GGUF uses tokenizer.ggml.add_eos_token = false
malicious artifact flips that metadata to true
ordinary prompt tokenization then silently appends the EOS token id

Reproduction

python3 verify_remote_poc.py --repo hacnho/gguf-add-eos-token-poc

Expected output delta

Prompt:

hello world

Control token IDs:

[1, 22172, 3186]

Malicious token IDs:

[1, 22172, 3186, 2]

The same EOS-appending behavior also appears on:

hello world
hello

Scanner posture

Local scout observed:

modelscan: No issues found on the malicious artifact

Downloads last month: 123

GGUF

Model size

0 params

Architecture

llama

Hardware compatibility

We're not able to determine the quantization variants.

View all variants

Inference Providers NEW

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