README.md

---
license: mit
license_link: https://huggingface.co/microsoft/phi-4/resolve/main/LICENSE
language:
- en
pipeline_tag: text-generation
base_model: microsoft/phi-4
tags:
- phi
- nlp
- math
- code
- chat
- conversational
- neuralmagic
- redhat
- llmcompressor
- quantized
- int8
---

# phi-4-quantized.w8a8

## Model Overview
- **Model Architecture:** Phi3ForCausalLM
  - **Input:** Text
  - **Output:** Text
- **Model Optimizations:**
  - **Activation quantization:** INT8
  - **Weight quantization:** INT8
- **Intended Use Cases:** This model is designed to accelerate research on language models, for use as a building block for generative AI powered features. It provides uses for general purpose AI systems and applications (primarily in English) which require:
  1. Memory/compute constrained environments.
  2. Latency bound scenarios.
  3. Reasoning and logic.
- **Out-of-scope:** This model is not specifically designed or evaluated for all downstream purposes, thus:
  1. Developers should consider common limitations of language models as they select use cases, and evaluate and mitigate for accuracy, safety, and fairness before using within a specific downstream use case, particularly for high-risk scenarios.
  2. Developers should be aware of and adhere to applicable laws or regulations (including privacy, trade compliance laws, etc.) that are relevant to their use case, including the model’s focus on English.
  3. Nothing contained in this Model Card should be interpreted as or deemed a restriction or modification to the license the model is released under.
- **Release Date:** 03/03/2025
- **Version:** 1.0
- **Model Developers:** Red Hat (Neural Magic)


### Model Optimizations

This model was obtained by quantizing activations and weights of [phi-4](https://huggingface.co/microsoft/phi-4) to INT8 data type.
This optimization reduces the number of bits used to represent weights and activations from 16 to 8, reducing GPU memory requirements (by approximately 50%) and increasing matrix-multiply compute throughput (by approximately 2x).
Weight quantization also reduces disk size requirements by approximately 50%.

Only weights and activations of the linear operators within transformers blocks are quantized.
Weights are quantized with a symmetric static per-channel scheme, whereas activations are quantized with a symmetric dynamic per-token scheme.
A combination of the [SmoothQuant](https://arxiv.org/abs/2211.10438) and [GPTQ](https://arxiv.org/abs/2210.17323) algorithms is applied for quantization, as implemented in the [llm-compressor](https://github.com/vllm-project/llm-compressor) library.


## Deployment

This model can be deployed efficiently using the [vLLM](https://docs.vllm.ai/en/latest/) backend, as shown in the example below.

```python
from vllm import LLM, SamplingParams
from transformers import AutoTokenizer

model_id = "neuralmagic-ent/phi-4-quantized.w8a8"
number_gpus = 1

sampling_params = SamplingParams(temperature=0.7, top_p=0.8, max_tokens=256)

tokenizer = AutoTokenizer.from_pretrained(model_id)

messages = [
    {"role": "user", "content": "Give me a short introduction to large language model."},
]

prompts = tokenizer.apply_chat_template(messages, tokenize=False)

llm = LLM(model=model_id, tensor_parallel_size=number_gpus)

outputs = llm.generate(prompts, sampling_params)

generated_text = outputs[0].outputs[0].text
print(generated_text)
```

vLLM aslo supports OpenAI-compatible serving. See the [documentation](https://docs.vllm.ai/en/latest/) for more details.

## Creation

<details>
  <summary>Creation details</summary>
  This model was created with [llm-compressor](https://github.com/vllm-project/llm-compressor) by running the code snippet below. 


  ```python
  from transformers import AutoModelForCausalLM, AutoTokenizer
  from llmcompressor.modifiers.quantization import GPTQModifier
  from llmcompressor.modifiers.smoothquant import SmoothQuantModifier
  from llmcompressor.transformers import oneshot
  from datasets import load_dataset

  # Load model
  model_stub = "microsoft/phi-4"
  model_name = model_stub.split("/")[-1]
  
  num_samples = 1024
  max_seq_len = 8192
  
  tokenizer = AutoTokenizer.from_pretrained(model_stub)
  
  model = AutoModelForCausalLM.from_pretrained(
      model_stub,
      device_map="auto",
      torch_dtype="auto",
  )
  
  def preprocess_fn(example):
    return {"text": tokenizer.apply_chat_template(example["messages"], add_generation_prompt=False, tokenize=False)}
  
  ds = load_dataset("neuralmagic/LLM_compression_calibration", split="train")
  ds = ds.map(preprocess_fn)
  
  # Configure the quantization algorithm and scheme
  recipe = [
      SmoothQuantModifier(
          smoothing_strength=0.7,
          mappings=[
              [["re:.*qkv_proj"], "re:.*input_layernorm"],
              [["re:.*gate_up_proj"], "re:.*post_attention_layernorm"],
          ],
      ),
      GPTQModifier(
          ignore=["lm_head"],
          sequential_targets=["Phi3DecoderLayer"],
          dampening_frac=0.01,
          targets="Linear",
          scheme="W8A8",
      ),
  ]
  
  # Apply quantization
  oneshot(
      model=model,
      dataset=ds, 
      recipe=recipe,
      max_seq_length=max_seq_len,
      num_calibration_samples=num_samples,
  )
  
  # Save to disk in compressed-tensors format
  save_path = model_name + "-quantized.w8a8"
  model.save_pretrained(save_path)
  tokenizer.save_pretrained(save_path)
  print(f"Model and tokenizer saved to: {save_path}")
  ```
</details>
 


## Evaluation

The model was evaluated on the OpenLLM leaderboard tasks (version 1) with the [lm-evaluation-harness](https://github.com/EleutherAI/lm-evaluation-harness) and the [vLLM](https://docs.vllm.ai/en/stable/) engine, using the following command:
```
lm_eval \
  --model vllm \
  --model_args pretrained="neuralmagic-ent/phi-4-quantized.w8a8",dtype=auto,gpu_memory_utilization=0.6,max_model_len=4096,enable_chunk_prefill=True,tensor_parallel_size=1 \
  --tasks openllm \
  --batch_size auto
```

### Accuracy

#### Open LLM Leaderboard evaluation scores
<table>
  <tr>
   <td><strong>Benchmark</strong>
   </td>
   <td><strong>phi-4</strong>
   </td>
   <td><strong>phi-4-quantized.w8a8<br>(this model)</strong>
   </td>
   <td><strong>Recovery</strong>
   </td>
  </tr>
  <tr>
   <td>MMLU (5-shot)
   </td>
   <td>80.30
   </td>
   <td>80.39
   </td>
   <td>100.1%
   </td>
  </tr>
  <tr>
   <td>ARC Challenge (25-shot)
   </td>
   <td>64.42
   </td>
   <td>64.33
   </td>
   <td>99.9%
   </td>
  </tr>
  <tr>
   <td>GSM-8K (5-shot, strict-match)
   </td>
   <td>90.07
   </td>
   <td>90.30
   </td>
   <td>100.3%
   </td>
  </tr>
  <tr>
   <td>Hellaswag (10-shot)
   </td>
   <td>84.37
   </td>
   <td>84.30
   </td>
   <td>99.9%
   </td>
  </tr>
  <tr>
   <td>Winogrande (5-shot)
   </td>
   <td>80.58
   </td>
   <td>79.95
   </td>
   <td>99.2%
   </td>
  </tr>
  <tr>
   <td>TruthfulQA (0-shot, mc2)
   </td>
   <td>59.37
   </td>
   <td>58.82
   </td>
   <td>99.1%
   </td>
  </tr>
  <tr>
   <td><strong>Average</strong>
   </td>
   <td><strong>76.52</strong>
   </td>
   <td><strong>76.35</strong>
   </td>
   <td><strong>99.8%</strong>
   </td>
  </tr>
</table>