Latest Content
Phi 4 icon

Phi 4

NVIDIA
Phi-4 is a dense transformer model designed for high-quality reasoning and efficient deployment in constrained environments. It features 14B parameters with 40 layers, 5,120 hidden size, and 40 attention heads with 10 key-value heads. The model supports a 16K token context window and is trained on 9.8T tokens using a mix of synthetic and curated data. It is optimized with supervised fine-tuning and direct preference optimization for strong instruction-following and low-latency inference.
TypeDense LLM
CapabilitiesText Generation, Instruction Following, Reasoning, Mathematical Reasoning+2 more
Group Release DateDecember 11, 2024
Links
Paper|Blog|HF Model Card
LicenseMIT

Inference Instructions

Deploy and run this model on NVIDIA B200 GPUs using the command below. Copy the command to get started with inference.

CONSOLE
docker run -it --rm 
 --runtime=nvidia 
 --gpus all 
 --ipc=host 
 --shm-size=128g 
 -p 8000:8000 
 -v ~/.cache/huggingface:/root/.cache/huggingface 
 -e HF_TOKEN='YOUR_HF_TOKEN' 
 vllm/vllm-openai:v0.18.1 
 microsoft/phi-4 
  --tensor-parallel-size 2 
 --max-model-len auto 
  --gpu-memory-utilization 0.95 
 --max-num-batched-tokens 65536 
 --max-num-seqs 1024 
 --trust-remote-code

Model Benchmarks

Each model was tested with a fixed input size and total token volume while increasing concurrency to measure serving performance under load.

ITL vs Concurrency

Time to First Token

Throughput Scaling

Total Tokens/sec vs Avg TTFT

Vultr Cloud GPU

NVIDIA HGX B200

Deploy NVIDIA B200 on Vultr Cloud GPU

Deploy NVIDIA B200 instancesContact Sales