GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
amd
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
amd
Model Tree
microsoft/Phi-4-reasoning-plus
Input Modalities
Output Modalities
Supported Functionality
This is a quantized version of Phi-4-reasoning-plus created by AMD using LLM Compressor (compressed-tensors) for ZenDNN-optimized CPU inference.
The model was quantized from Phi-4-reasoning-plus using LLM Compressor via the GPTQ algorithm.
compressed-tensors, num_bits=4, type=int, symmetric=true, group_size=128group_size=128, actorder=static), stored as pack-quantizedlm_headpython
import torchfrom datasets import load_datasetfrom transformers import AutoModelForCausalLM, AutoTokenizerfrom llmcompressor import oneshotfrom llmcompressor.modifiers.gptq import GPTQModifiermodel_id = "microsoft/Phi-4-reasoning-plus"output_dir = "./Phi-4-reasoning-plus-w4a16-llmcompressor-v0.12.0"NUM_CALIBRATION_SAMPLES = 128MAX_SEQUENCE_LENGTH = 2048# Step 1: Load the BF16 model and tokenizermodel = AutoModelForCausalLM.from_pretrained(model_id,dtype=torch.bfloat16,device_map="cpu",trust_remote_code=True,)tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True)# Step 2: Load calibration data. GPTQ is data-driven: it needs real activations# to build the per-layer Hessians used to compensate the rounding error.ds = load_dataset("HuggingFaceH4/ultrachat_200k",split=f"train_sft[:{NUM_CALIBRATION_SAMPLES}]",)ds = ds.map(lambda example: {"text": "\n".join(m["content"] for m in example["messages"])},remove_columns=ds.column_names,)# Step 3: Define the W4A16 GPTQ reciperecipe = GPTQModifier(scheme="W4A16", targets="Linear", ignore=["lm_head"])# Step 4: One-shot quantize with calibration and save in compressed-tensors formatoneshot(model=model,dataset=ds,recipe=recipe,max_seq_length=MAX_SEQUENCE_LENGTH,tokenizer=tokenizer,output_dir=output_dir,trust_remote_code_model=True,)# Smoke testinputs = tokenizer("What are we having for dinner?", return_tensors="pt")output = model.generate(**inputs, max_new_tokens=30)print(tokenizer.decode(output[0], skip_special_tokens=True))
python
from vllm import LLM, SamplingParamsmodel = LLM(model="amd/Phi-4-reasoning-plus-w4a16-llmcompressor-v0.12.0",dtype="bfloat16",)sampling_params = SamplingParams(temperature=0.7, max_tokens=256)outputs = model.generate(["Hello, how are you?"], sampling_params)print(outputs[0].outputs[0].text)
markdown
torch==2.11.0zentorch==2.11.0.3vllm==0.26.0llmcompressor==0.12.0
For optimal performance, set LD_PRELOAD with libomp.so (LLVM OpenMP) or libiomp5.so (Intel OpenMP):
bash
# Using LLVM OpenMP (llvmopenmp)export LD_PRELOAD=$(find /path/to/env -name "libomp.so" | head -1)# Or using Intel OpenMP (libiomp)export LD_PRELOAD=$(find /path/to/env -name "libiomp5.so" | head -1)
Note: Set
LD_PRELOADbefore launching vLLM or any inference script.
The model was evaluated against the BF16 (unquantized) baseline on standard benchmarks using lm-evaluation-harness with the vLLM engine.
| Benchmark | BF16 Baseline | W4A16 (this model) | Recovery |
|---|---|---|---|
| GSM8K (5-shot) | 0.8704 | 0.8688 | 99.82% |
bash
lm_eval \--model vllm \--model_args pretrained=amd/Phi-4-reasoning-plus-w4a16-llmcompressor-v0.12.0,dtype=bfloat16 \--tasks gsm8k \--batch_size auto \--trust_remote_code \--num_fewshot 5 \--apply_chat_template \--log_samples \--gen_kwargs "max_gen_toks=2048" \--output_path .
This model is distributed under the same license as the source model. See the LICENSE file for details.
Modifications copyright (c) 2026 Advanced Micro Devices, Inc. All rights reserved.