Phi-4 Technical Report
Paper • 2412.08905 • Published • 123
How to use vincentoh/phi-4_f16_ollama with llama.cpp:
curl -LsSf https://llama.app/install.sh | sh # Start a local OpenAI-compatible server with a web UI: llama serve -hf vincentoh/phi-4_f16_ollama:F16 # Run inference directly in the terminal: llama cli -hf vincentoh/phi-4_f16_ollama:F16
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf vincentoh/phi-4_f16_ollama:F16 # Run inference directly in the terminal: llama cli -hf vincentoh/phi-4_f16_ollama:F16
# 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 vincentoh/phi-4_f16_ollama:F16 # Run inference directly in the terminal: ./llama-cli -hf vincentoh/phi-4_f16_ollama:F16
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 vincentoh/phi-4_f16_ollama:F16 # Run inference directly in the terminal: ./build/bin/llama-cli -hf vincentoh/phi-4_f16_ollama:F16
docker model run hf.co/vincentoh/phi-4_f16_ollama:F16
How to use vincentoh/phi-4_f16_ollama with Ollama:
ollama run hf.co/vincentoh/phi-4_f16_ollama:F16
How to use vincentoh/phi-4_f16_ollama with Docker Model Runner:
docker model run hf.co/vincentoh/phi-4_f16_ollama:F16
How to use vincentoh/phi-4_f16_ollama with Lemonade:
# Download Lemonade from https://lemonade-server.ai/ lemonade pull vincentoh/phi-4_f16_ollama:F16
lemonade run user.phi-4_f16_ollama-F16
lemonade list
for testing phi4 against gpqa_diamond.csv dataset
example script here created by phi4: https://gist.github.com/bigsnarfdude/7754cfb69f94d540da7bda7ef6a910ae
Microsoft technical report: https://arxiv.org/pdf/2412.08905
requires 30GB VRAM
16-bit