Instructions to use mozilla-ai/llava-v1.5-7b-llamafile with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use mozilla-ai/llava-v1.5-7b-llamafile 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 mozilla-ai/llava-v1.5-7b-llamafile:Q8_0 # Run inference directly in the terminal: llama cli -hf mozilla-ai/llava-v1.5-7b-llamafile:Q8_0
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf mozilla-ai/llava-v1.5-7b-llamafile:Q8_0 # Run inference directly in the terminal: llama cli -hf mozilla-ai/llava-v1.5-7b-llamafile:Q8_0
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 mozilla-ai/llava-v1.5-7b-llamafile:Q8_0 # Run inference directly in the terminal: ./llama-cli -hf mozilla-ai/llava-v1.5-7b-llamafile:Q8_0
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 mozilla-ai/llava-v1.5-7b-llamafile:Q8_0 # Run inference directly in the terminal: ./build/bin/llama-cli -hf mozilla-ai/llava-v1.5-7b-llamafile:Q8_0
Use Docker
docker model run hf.co/mozilla-ai/llava-v1.5-7b-llamafile:Q8_0
- LM Studio
- Jan
- Ollama
How to use mozilla-ai/llava-v1.5-7b-llamafile with Ollama:
ollama run hf.co/mozilla-ai/llava-v1.5-7b-llamafile:Q8_0
- Unsloth Studio
How to use mozilla-ai/llava-v1.5-7b-llamafile 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 mozilla-ai/llava-v1.5-7b-llamafile 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 mozilla-ai/llava-v1.5-7b-llamafile to start chatting
Using HuggingFace Spaces for Unsloth
# No setup required # Open https://huggingface.co/spaces/unsloth/studio in your browser # Search for mozilla-ai/llava-v1.5-7b-llamafile to start chatting
- Docker Model Runner
How to use mozilla-ai/llava-v1.5-7b-llamafile with Docker Model Runner:
docker model run hf.co/mozilla-ai/llava-v1.5-7b-llamafile:Q8_0
- Lemonade
How to use mozilla-ai/llava-v1.5-7b-llamafile with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull mozilla-ai/llava-v1.5-7b-llamafile:Q8_0
Run and chat with the model
lemonade run user.llava-v1.5-7b-llamafile-Q8_0
List all available models
lemonade list
- Atomic Chat
llamafile: virus message.
The llamafile can not be downloaded because windows blocks it due to a virus message. Please advise, thank you.
Similar issue, Bitdefender just quarantined it.
if you don't trust it you can download the model and llamafile separately
Virus scanners sometimes get upset about our binaries that have weights inside them. We can't get them unflagged, because the online form with Microsoft Security Intelligence has an upload limit. What we can promise, is that the llamafile release binaries won't be flagged by Windows Defender. (We have zero support for any other AVs). So, could you try these commands:
curl -L -o llamafile.exe https://github.com/Mozilla-Ocho/llamafile/releases/download/0.7/llamafile-0.7
curl -L -o llava-v1.5-7b-Q4_K.gguf https://huggingface.co/jartine/llava-v1.5-7B-GGUF/resolve/main/llava-v1.5-7b-Q4_K.gguf
curl -L -o llava-v1.5-7b-mmproj-Q8_0.gguf https://huggingface.co/jartine/llava-v1.5-7B-GGUF/resolve/main/llava-v1.5-7b-mmproj-Q8_0.gguf
.\llamafile.exe -m llava-v1.5-7b-Q4_K.gguf --mmproj llava-v1.5-7b-mmproj-Q8_0.gguf --image SOMETHING.JPG -p "### User: What do you see?\n### Assistant:"
If that still gets flagged by Windows Defender, you should file an issue on https://github.com/mozilla-ocho/llamafile/