Instructions to use tiny-random/phi-4-flash with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use tiny-random/phi-4-flash with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="tiny-random/phi-4-flash", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("tiny-random/phi-4-flash", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use tiny-random/phi-4-flash with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "tiny-random/phi-4-flash"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tiny-random/phi-4-flash",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/tiny-random/phi-4-flash

SGLang

How to use tiny-random/phi-4-flash with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "tiny-random/phi-4-flash" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tiny-random/phi-4-flash",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "tiny-random/phi-4-flash" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tiny-random/phi-4-flash",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use tiny-random/phi-4-flash with Docker Model Runner:
```
docker model run hf.co/tiny-random/phi-4-flash
```

phi-4-flash

File size: 1,120 Bytes

0c24726

{
  "architectures": [
    "Phi4FlashForCausalLM"
  ],
  "attention_dropout": 0.0,
  "auto_map": {
    "AutoConfig": "microsoft/Phi-4-mini-flash-reasoning--configuration_phi4flash.Phi4FlashConfig",
    "AutoModelForCausalLM": "microsoft/Phi-4-mini-flash-reasoning--modeling_phi4flash.Phi4FlashForCausalLM",
    "AutoTokenizer": "Xenova/gpt-4o"
  },
  "bos_token_id": 199999,
  "embd_pdrop": 0.0,
  "eos_token_id": 199999,
  "hidden_act": "silu",
  "hidden_size": 64,
  "initializer_range": 0.02,
  "intermediate_size": 64,
  "layer_norm_eps": 1e-05,
  "lm_head_bias": false,
  "mamba_conv_bias": true,
  "mamba_d_conv": 4,
  "mamba_d_state": 16,
  "mamba_dt_rank": 4,
  "mamba_expand": 2,
  "mamba_proj_bias": false,
  "max_position_embeddings": 262144,
  "mb_per_layer": 2,
  "mlp_bias": false,
  "model_type": "phi4flash",
  "num_attention_heads": 2,
  "num_hidden_layers": 4,
  "num_key_value_heads": 2,
  "pad_token_id": 199999,
  "resid_pdrop": 0.0,
  "rope_theta": 10000.0,
  "sliding_window": 512,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.54.0.dev0",
  "use_cache": true,
  "vocab_size": 200064
}