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

Libraries

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

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

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

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

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

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

Install from pip and serve model

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

Use Docker

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

SGLang

How to use yujiepan/phi-4-flash-tiny-random 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 "yujiepan/phi-4-flash-tiny-random" \
    --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": "yujiepan/phi-4-flash-tiny-random",
		"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 "yujiepan/phi-4-flash-tiny-random" \
        --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": "yujiepan/phi-4-flash-tiny-random",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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

phi-4-flash-tiny-random / config.json

yujiepan

Upload folder using huggingface_hub

da2d373 verified 10 months ago

raw

history blame contribute delete

1.12 kB

	{
	"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
	}