Instructions to use pytorch/Phi-4-mini-instruct-INT8-INT4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use pytorch/Phi-4-mini-instruct-INT8-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-INT8-INT4", 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("pytorch/Phi-4-mini-instruct-INT8-INT4", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("pytorch/Phi-4-mini-instruct-INT8-INT4", trust_remote_code=True)
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 pytorch/Phi-4-mini-instruct-INT8-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-INT8-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-INT8-INT4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/pytorch/Phi-4-mini-instruct-INT8-INT4

SGLang

How to use pytorch/Phi-4-mini-instruct-INT8-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-INT8-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-INT8-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-INT8-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-INT8-INT4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use pytorch/Phi-4-mini-instruct-INT8-INT4 with Docker Model Runner:
```
docker model run hf.co/pytorch/Phi-4-mini-instruct-INT8-INT4
```

Update README.md

by metascroy - opened Apr 30, 2025

base: refs/heads/main

←

from: refs/pr/4

Discussion Files changed

+20

-11

Files changed (1) hide show

README.md +20 -11

README.md CHANGED Viewed

@@ -30,24 +30,33 @@ We used following code to get the quantized model:
 ```
 from transformers import (
-    AutoModelForCausalLM,
-    AutoProcessor,
-    AutoTokenizer,
-    TorchAoConfig,
 )
 from torchao.quantization.quant_api import (
     Int8DynamicActivationIntxWeightConfig,
 )
-from torchao.quantization.granularity import PerGroup
 import torch
 model_id = "microsoft/Phi-4-mini-instruct"
 linear_config = Int8DynamicActivationIntxWeightConfig(
     weight_dtype=torch.int4,
     weight_granularity=PerGroup(32),
 )
-quantization_config = TorchAoConfig(quant_type=linear_config)
-quantized_model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype="auto", device_map="auto", quantization_config=quantization_config)
 tokenizer = AutoTokenizer.from_pretrained(model_id)
 # Push to hub
@@ -91,7 +100,7 @@ torch.save(state_dict, "phi4-mini-8dq4w.pt")
 The response from the manual testing is:
 ```
-Hello! As an AI, I don't have consciousness in the way humans do, but I'm here and ready to assist you. How can I help you today?
 ```
 # Model Quality
@@ -100,7 +109,7 @@ We rely on [lm-evaluation-harness](https://github.com/EleutherAI/lm-evaluation-h
 ## baseline
 ```
-lm_eval --model hf --model_args pretrained=microsoft/Phi-4-mini-instruct --tasks hellaswag --device cuda:0 --batch_size 8
 ```
 ## 8dq4w
@@ -111,7 +120,7 @@ from lm_eval.utils import (
     make_table,
 )
-lm_eval_model = lm_eval.models.huggingface.HFLM(pretrained=quantized_model, batch_size=8)
 results = evaluator.simple_evaluate(
     lm_eval_model, tasks=["hellaswag"], device="cuda:0", batch_size="auto"
 )
@@ -123,7 +132,7 @@ print(make_table(results))
 |                                  | Phi-4 mini-Ins | phi4-mini-8dq4w |
 | **Popular aggregated benchmark** |             |                   |
 | **Reasoning**                    |             |                   |
-| HellaSwag                        | 54.57        | 53.19            |
 | **Multilingual**                 |             |                   |
 | **Math**                         |             |                   |
 | **Overall**                      | **TODO**    | **TODO**          |

 ```
 from transformers import (
+  AutoModelForCausalLM,
+  AutoProcessor,
+  AutoTokenizer,
+  TorchAoConfig,
 )
 from torchao.quantization.quant_api import (
+    IntxWeightOnlyConfig,
     Int8DynamicActivationIntxWeightConfig,
+    AOPerModuleConfig
 )
+from torchao.quantization.granularity import PerGroup, PerAxis
 import torch
 model_id = "microsoft/Phi-4-mini-instruct"
+embedding_config = IntxWeightOnlyConfig(
+    weight_dtype=torch.int8,
+    granularity=PerAxis(0),
+)
 linear_config = Int8DynamicActivationIntxWeightConfig(
     weight_dtype=torch.int4,
     weight_granularity=PerGroup(32),
+    weight_scale_dtype=torch.bfloat16,
 )
+quant_config = AOPerModuleConfig({"_default": linear_config, "model.embed_tokens": embedding_config})
+quantization_config = TorchAoConfig(quant_type=quant_config, include_embedding=True)
+quantized_model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype=torch.float32, device_map="auto", quantization_config=quantization_config)
 tokenizer = AutoTokenizer.from_pretrained(model_id)
 # Push to hub
 The response from the manual testing is:
 ```
+Hello! As an AI, I don't have consciousness in the way humans do, but I am fully operational and here to assist you. How can I help you today?
 ```
 # Model Quality
 ## baseline
 ```
+lm_eval --model hf --model_args pretrained=microsoft/Phi-4-mini-instruct --tasks hellaswag --device cuda:0 --batch_size 64
 ```
 ## 8dq4w
     make_table,
 )
+lm_eval_model = lm_eval.models.huggingface.HFLM(pretrained=quantized_model, batch_size=64)
 results = evaluator.simple_evaluate(
     lm_eval_model, tasks=["hellaswag"], device="cuda:0", batch_size="auto"
 )
 |                                  | Phi-4 mini-Ins | phi4-mini-8dq4w |
 | **Popular aggregated benchmark** |             |                   |
 | **Reasoning**                    |             |                   |
+| HellaSwag                        | 54.57        | 53.24            |
 | **Multilingual**                 |             |                   |
 | **Math**                         |             |                   |
 | **Overall**                      | **TODO**    | **TODO**          |