Instructions to use Pinkstack/Phi-4-mini-6b-merge with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Pinkstack/Phi-4-mini-6b-merge with Transformers:

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

pipe = pipeline("text-generation", model="Pinkstack/Phi-4-mini-6b-merge", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Pinkstack/Phi-4-mini-6b-merge", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("Pinkstack/Phi-4-mini-6b-merge", trust_remote_code=True, device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Pinkstack/Phi-4-mini-6b-merge with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Pinkstack/Phi-4-mini-6b-merge"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Pinkstack/Phi-4-mini-6b-merge",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Pinkstack/Phi-4-mini-6b-merge

SGLang

How to use Pinkstack/Phi-4-mini-6b-merge 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 "Pinkstack/Phi-4-mini-6b-merge" \
    --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": "Pinkstack/Phi-4-mini-6b-merge",
		"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 "Pinkstack/Phi-4-mini-6b-merge" \
        --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": "Pinkstack/Phi-4-mini-6b-merge",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Pinkstack/Phi-4-mini-6b-merge with Docker Model Runner:
```
docker model run hf.co/Pinkstack/Phi-4-mini-6b-merge
```

Pinkstack commited on May 28, 2025

Commit

1050914

verified ·

1 Parent(s): f817e22

Update sample_finetune.py

Browse files

Files changed (1) hide show

sample_finetune.py +2 -2

sample_finetune.py CHANGED Viewed

@@ -128,10 +128,10 @@ logger.info(f"PEFT parameters {peft_conf}")
 ################
 # Model Loading
 ################
-checkpoint_path = "microsoft/Phi-4-mini-instruct"
 model_kwargs = dict(
     use_cache=False,
-    trust_remote_code=True,
     attn_implementation="flash_attention_2",  # loading the model with flash-attention support
     torch_dtype=torch.bfloat16,
     device_map=None

 ################
 # Model Loading
 ################
+checkpoint_path = "Pinkstack/Phi-4-mini-6b-merge"
 model_kwargs = dict(
     use_cache=False,
+    trust_remote_code=False,
     attn_implementation="flash_attention_2",  # loading the model with flash-attention support
     torch_dtype=torch.bfloat16,
     device_map=None