Instructions to use bartowski/QVQ-72B-Preview-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- llama-cpp-python
How to use bartowski/QVQ-72B-Preview-GGUF with llama-cpp-python:
# !pip install llama-cpp-python from llama_cpp import Llama llm = Llama.from_pretrained( repo_id="bartowski/QVQ-72B-Preview-GGUF", filename="QVQ-72B-Preview-IQ1_M.gguf", )
llm.create_chat_completion( messages = [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] ) - Notebooks
- Google Colab
- Kaggle
- Local Apps
- llama.cpp
How to use bartowski/QVQ-72B-Preview-GGUF with llama.cpp:
Install from brew
brew install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama-server -hf bartowski/QVQ-72B-Preview-GGUF:Q4_K_M # Run inference directly in the terminal: llama-cli -hf bartowski/QVQ-72B-Preview-GGUF:Q4_K_M
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama-server -hf bartowski/QVQ-72B-Preview-GGUF:Q4_K_M # Run inference directly in the terminal: llama-cli -hf bartowski/QVQ-72B-Preview-GGUF: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 bartowski/QVQ-72B-Preview-GGUF:Q4_K_M # Run inference directly in the terminal: ./llama-cli -hf bartowski/QVQ-72B-Preview-GGUF: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 bartowski/QVQ-72B-Preview-GGUF:Q4_K_M # Run inference directly in the terminal: ./build/bin/llama-cli -hf bartowski/QVQ-72B-Preview-GGUF:Q4_K_M
Use Docker
docker model run hf.co/bartowski/QVQ-72B-Preview-GGUF:Q4_K_M
- LM Studio
- Jan
- vLLM
How to use bartowski/QVQ-72B-Preview-GGUF with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "bartowski/QVQ-72B-Preview-GGUF" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "bartowski/QVQ-72B-Preview-GGUF", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Describe this image in one sentence." }, { "type": "image_url", "image_url": { "url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg" } } ] } ] }'Use Docker
docker model run hf.co/bartowski/QVQ-72B-Preview-GGUF:Q4_K_M
- Ollama
How to use bartowski/QVQ-72B-Preview-GGUF with Ollama:
ollama run hf.co/bartowski/QVQ-72B-Preview-GGUF:Q4_K_M
- Unsloth Studio new
How to use bartowski/QVQ-72B-Preview-GGUF 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 bartowski/QVQ-72B-Preview-GGUF 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 bartowski/QVQ-72B-Preview-GGUF to start chatting
Using HuggingFace Spaces for Unsloth
# No setup required # Open https://huggingface.co/spaces/unsloth/studio in your browser # Search for bartowski/QVQ-72B-Preview-GGUF to start chatting
- Docker Model Runner
How to use bartowski/QVQ-72B-Preview-GGUF with Docker Model Runner:
docker model run hf.co/bartowski/QVQ-72B-Preview-GGUF:Q4_K_M
- Lemonade
How to use bartowski/QVQ-72B-Preview-GGUF with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull bartowski/QVQ-72B-Preview-GGUF:Q4_K_M
Run and chat with the model
lemonade run user.QVQ-72B-Preview-GGUF-Q4_K_M
List all available models
lemonade list
Ollama upload please.
Hey, thanks so much, I really appreciate it! Would you mind uploading it to Ollama? I can do it too, but downloading the model and then uploading it again would take me quite a while.
I was going to say you can run directly in ollama from HF but I actually don't know how that works with vision models.. will look into it !
thank you!
I was going to say you can run directly in ollama from HF but I actually don't know how that works with vision models.. will look into it !
well i imported the gguff into openwebui and....
this happens every time and here is the translation:
, Hello little fairies~
Your editor is online again
Recently I'm following "The Next Stop Is Happiness"
It's really too sweet
Song Weilong and Song Qian have a full sense of CP
wait... this is diffrent a while ago it was yapping about a cook and giving me recipies
Hi, seems like we have a problem parsing GGUF file on hugging face backend. I'll report it to the team
So, this is a real bug? I thought this was a problem on my side, thanks a lot! Now I don't have to debug my Ollama or OWUI.
Yes it's a bug on HF side. I don't know yet when we can deploy a fix, given that many of us are on vacation.
In the meantime, you can try adding the chat template yourself via a Modelfile, ref ollama docs
FROM hf.co/bartowski/QVQ-72B-Preview-GGUF
TEMPLATE """{{ if .System }}<|im_start|>system
{{ .System }}<|im_end|>
{{ end }}{{ if .Prompt }}<|im_start|>user
{{ .Prompt }}<|im_end|>
{{ end }}<|im_start|>assistant
"""
PARAMETER stop "<|im_end|>"
thank you and also, is the problem with all the ggufs or just the q4km's?
it's for all quants
any updates?
I'll try to take a look tomorrow at the process
❤️ thanks a lot! you are a legend!
Haven't been able to figure it out :( was able to upload a file, but not sure how vision is meant to be uploaded/work, if anyone has insight feel free to let me know but the documentation seems quite lacking
Hi @bartowski , sorry for the late response. It seems like the fix can be applied from HF server side. I'll discuss this with our team this week.
It should work correctly now. Tested using https://huggingface.co/spaces/ngxson/debug_ollama_manifest
======================
Template:
{{ if .System }}<|im_start|>system
{{ .System }}<|im_end|>
{{ end }}{{ if .Prompt }}<|im_start|>user
{{ .Prompt }}<|im_end|>
{{ end }}<|im_start|>assistant
{{ .Response }}<|im_end|>
======================
Parameters:
{
"stop": [
"<|im_start|>",
"<|im_end|>",
"<|im_start|>",
"<|im_end|>",
"<|im_start|>",
"<|im_end|>",
"<|im_start|>user"
]
}
so i just have redownload and import it and it should work?