Instructions to use pytorch/Phi-4-mini-instruct-AWQ-INT4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use pytorch/Phi-4-mini-instruct-AWQ-INT4 with Transformers:

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

pipe = pipeline("text-generation", model="pytorch/Phi-4-mini-instruct-AWQ-INT4", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("pytorch/Phi-4-mini-instruct-AWQ-INT4", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("pytorch/Phi-4-mini-instruct-AWQ-INT4", 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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use pytorch/Phi-4-mini-instruct-AWQ-INT4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "pytorch/Phi-4-mini-instruct-AWQ-INT4"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "pytorch/Phi-4-mini-instruct-AWQ-INT4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/pytorch/Phi-4-mini-instruct-AWQ-INT4

SGLang

How to use pytorch/Phi-4-mini-instruct-AWQ-INT4 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 "pytorch/Phi-4-mini-instruct-AWQ-INT4" \
    --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": "pytorch/Phi-4-mini-instruct-AWQ-INT4",
		"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 "pytorch/Phi-4-mini-instruct-AWQ-INT4" \
        --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": "pytorch/Phi-4-mini-instruct-AWQ-INT4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use pytorch/Phi-4-mini-instruct-AWQ-INT4 with Docker Model Runner:
```
docker model run hf.co/pytorch/Phi-4-mini-instruct-AWQ-INT4
```

jerryzh168 commited on Sep 15, 2025

Commit

a2f67ff

verified ·

1 Parent(s): 32d06e3

Update README.md

Browse files

Files changed (1) hide show

README.md +6 -5

README.md CHANGED Viewed

@@ -223,7 +223,7 @@ Since the checkpoint is tuned on `mmlu_pro`, we check against the accuracy for `
 | Benchmark                        |                |                           |                           |
 |----------------------------------|----------------|---------------------------|---------------------------|
 |                                  | microsoft/Phi-4-mini-instruct   | pytorch/Phi-4-mini-instruct-INT4 | pytorch/Phi-4-mini-instruct-AWQ-INT4
-| mmlu_pro                             | 46.43   | 36.74                      |    43.13        |
 <details>
@@ -313,13 +313,14 @@ print(f"Peak Memory Usage: {mem:.02f} GB")
 ## Results (H100 machine)
 | Benchmark (Latency)              |                |                          |                          |
 |----------------------------------|----------------|--------------------------|--------------------------|
-|                                  | microsoft/Phi-4-mini-instruct   | jerryzh168/Phi-4-mini-instruct-INT4 | pytorch/Phi-4-mini-instruct-AWQ-INT4
-| latency (batch_size=1)           | 1.60s          | TODO    | 1.37s (1.17x speedup) |
-| latency (batch_size=256)         | 5.47s          | TODO    | 5.55s (0.98x speedup) |
 Note: it's expected that the awq-int4 checkpoint is slower when batch size is 256 since the problem is not memory bound but becomes compute bound when batch size is larger, while
 int4 weight only checkpoint is only expected to have speedup for memory bound situations.
-Note: we are comparing to jerryzh168/Phi-4-mini-instruct-INT4 which is a checkpoint for H100, since the AWQ-INT4 is for the new INT4 config that's optimized for H100.
 <details>
 <summary> Reproduce Model Performance Results </summary>

 | Benchmark                        |                |                           |                           |
 |----------------------------------|----------------|---------------------------|---------------------------|
 |                                  | microsoft/Phi-4-mini-instruct   | pytorch/Phi-4-mini-instruct-INT4 | pytorch/Phi-4-mini-instruct-AWQ-INT4
+| mmlu_pro                         | 46.43   | 36.74                      |    43.13        |
 <details>
 ## Results (H100 machine)
 | Benchmark (Latency)              |                |                          |                          |
 |----------------------------------|----------------|--------------------------|--------------------------|
+|                                  | microsoft/Phi-4-mini-instruct | jerryzh168/Phi-4-mini-instruct-INT4 | pytorch/Phi-4-mini-instruct-AWQ-INT4
+| latency (batch_size=1)           | 1.61s          | 1.09s (1.48x speedup)    | 1.37s (1.17x speedup) |
+| latency (batch_size=256)         | 5.31s          | 32.33s    | 5.44s (0.98x speedup) |
 Note: it's expected that the awq-int4 checkpoint is slower when batch size is 256 since the problem is not memory bound but becomes compute bound when batch size is larger, while
 int4 weight only checkpoint is only expected to have speedup for memory bound situations.
+Note: we are comparing to jerryzh168/Phi-4-mini-instruct-INT4 which is a checkpoint for H100, since the AWQ-INT4 is using the new INT4 config that's optimized for H100. It's possible to generate
+AWQ-INT4 for A100 as well using `Int4WeightOnlyConfig(group_size=128, int4_packing_foramt="tile_packed_to_4d", int4_choose_qparams_algorithm="hqq")`
 <details>
 <summary> Reproduce Model Performance Results </summary>