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

Libraries

How to use microsoft/phi-4 with Transformers:

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

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

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("microsoft/phi-4")
model = AutoModelForCausalLM.from_pretrained("microsoft/phi-4")
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

vLLM

How to use microsoft/phi-4 with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/microsoft/phi-4

SGLang

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

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

gugarosa commited on Feb 4, 2025

Commit

6fbb3d3

verified ·

1 Parent(s): f957856

Suggested tokenizer changes by Unsloth.ai (#21)

Browse files

- fix(root): Adds proposed changes by unsloth.ai. (b15fb484864f7247bfec0860d31e105173e809c7)
- chore(root): Changes pad_token to <|dummy_85|>. (492fc706206120688c0cc9d538e6be3ad6558f7f)

Files changed (4) hide show

config.json +2 -2
generation_config.json +2 -4
special_tokens_map.json +2 -2
tokenizer_config.json +3 -3

config.json CHANGED Viewed

@@ -8,7 +8,7 @@
   "auto_map": {},
   "bos_token_id": 100257,
   "embd_pdrop": 0.0,
-  "eos_token_id": 100257,
   "hidden_act": "silu",
   "hidden_size": 5120,
   "initializer_range": 0.02,
@@ -19,7 +19,7 @@
   "num_hidden_layers": 40,
   "num_key_value_heads": 10,
   "original_max_position_embeddings": 16384,
-  "pad_token_id": 100257,
   "resid_pdrop": 0.0,
   "rms_norm_eps": 1e-05,
   "rope_scaling": null,

   "auto_map": {},
   "bos_token_id": 100257,
   "embd_pdrop": 0.0,
+  "eos_token_id": 100265,
   "hidden_act": "silu",
   "hidden_size": 5120,
   "initializer_range": 0.02,
   "num_hidden_layers": 40,
   "num_key_value_heads": 10,
   "original_max_position_embeddings": 16384,
+  "pad_token_id": 100349,
   "resid_pdrop": 0.0,
   "rms_norm_eps": 1e-05,
   "rope_scaling": null,

generation_config.json CHANGED Viewed

@@ -1,9 +1,7 @@
 {
     "_from_model_config": true,
     "bos_token_id": 100257,
-    "eos_token_id": [
-      100257,
-      100265
-    ],
     "transformers_version": "4.47.0"
 }

 {
     "_from_model_config": true,
     "bos_token_id": 100257,
+    "eos_token_id": 100265,
+    "pad_token_id": 100349,
     "transformers_version": "4.47.0"
 }

special_tokens_map.json CHANGED Viewed

@@ -1,5 +1,5 @@
 {
   "bos_token": "<|endoftext|>",
-  "eos_token": "<|endoftext|>",
-  "pad_token": "<|endoftext|>"
 }

 {
   "bos_token": "<|endoftext|>",
+  "eos_token": "<|im_end|>",
+  "pad_token": "<|dummy_85|>"
 }

tokenizer_config.json CHANGED Viewed

@@ -771,10 +771,10 @@
     }
   },
   "bos_token": "<|endoftext|>",
-  "chat_template": "{% for message in messages %}{% if (message['role'] == 'system') %}{{'<|im_start|>system<|im_sep|>' + message['content'] + '<|im_end|>'}}{% elif (message['role'] == 'user') %}{{'<|im_start|>user<|im_sep|>' + message['content'] + '<|im_end|><|im_start|>assistant<|im_sep|>'}}{% elif (message['role'] == 'assistant') %}{{message['content'] + '<|im_end|>'}}{% endif %}{% endfor %}",
   "clean_up_tokenization_spaces": false,
-  "eos_token": "<|endoftext|>",
   "model_max_length": 16384,
-  "pad_token": "<|endoftext|>",
   "tokenizer_class": "GPT2Tokenizer"
 }

     }
   },
   "bos_token": "<|endoftext|>",
+  "chat_template": "{% for message in messages %}{% if (message['role'] == 'system') %}{{'<|im_start|>system<|im_sep|>' + message['content'] + '<|im_end|>'}}{% elif (message['role'] == 'user') %}{{'<|im_start|>user<|im_sep|>' + message['content'] + '<|im_end|>'}}{% elif (message['role'] == 'assistant') %}{{'<|im_start|>assistant<|im_sep|>' + message['content'] + '<|im_end|>'}}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant<|im_sep|>' }}{% endif %}",
   "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
   "model_max_length": 16384,
+  "pad_token": "<|dummy_85|>",
   "tokenizer_class": "GPT2Tokenizer"
 }