Instructions to use VisualArchiveSystem/vas-models with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use VisualArchiveSystem/vas-models 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 VisualArchiveSystem/vas-models:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf VisualArchiveSystem/vas-models:Q4_K_M

Install from WinGet (Windows)

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

Use Docker

docker model run hf.co/VisualArchiveSystem/vas-models:Q4_K_M

LM Studio
Jan
Ollama
How to use VisualArchiveSystem/vas-models with Ollama:
```
ollama run hf.co/VisualArchiveSystem/vas-models:Q4_K_M
```

Unsloth Studio

How to use VisualArchiveSystem/vas-models 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 VisualArchiveSystem/vas-models 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 VisualArchiveSystem/vas-models to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for VisualArchiveSystem/vas-models to start chatting

Atomic Chat new
Docker Model Runner
How to use VisualArchiveSystem/vas-models with Docker Model Runner:
```
docker model run hf.co/VisualArchiveSystem/vas-models:Q4_K_M
```

Lemonade

How to use VisualArchiveSystem/vas-models with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull VisualArchiveSystem/vas-models:Q4_K_M

Run and chat with the model

lemonade run user.vas-models-Q4_K_M

List all available models

lemonade list

VAS Pro — AI Models

Pre-quantized GGUF models for VAS Pro (Visual Archive System) local AI assistant.

Models

Model	File	Size	Purpose	Original
Phi-4 Mini	`phi-4-mini-Q4_K_M.gguf`	~2.5 GB	Fast responses, greetings	microsoft/Phi-4-mini-instruct
Gemma 3 4B	`gemma-4-4b-it-Q4_K_M.gguf`	~2.5 GB	Standard tasks, tool calling	google/gemma-3-4b-it
Qwen2.5-VL 7B	`qwen3.5-9b-vision-Q4_K_M.gguf`	~4.7 GB	Vision, OCR, image analysis	Qwen/Qwen2.5-VL-7B-Instruct
MxBAI Embed Large	`mxbai-embed-large-v1-f16.gguf`	~670 MB	Semantic search embeddings	mixedbread-ai/mxbai-embed-large-v1

Usage

These models are automatically downloaded by VAS Pro on first run. No manual setup required.

Quantization

Text models use Q4_K_M quantization (best quality/size ratio for 4-bit)
Embedding model uses F16 (full precision for maximum retrieval accuracy)

License

Models retain their original licenses:

Phi-4 Mini: MIT License
Gemma 3: Gemma Terms of Use
Qwen2.5-VL: Apache 2.0
MxBAI Embed: Apache 2.0

Downloads last month: 33

GGUF

Model size

4B params

Architecture

gemma3

Hardware compatibility

4-bit

16-bit

Inference Providers NEW

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