intel-npu-llm/requirements.txt at main · prskid1000/intel-npu-llm · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
# ============================================================================
# OpenVINO GenAI API Server - All Dependencies
# Install everything: pip install -r requirements.txt
# ============================================================================

# Core API Server
fastapi>=0.104.0
uvicorn[standard]>=0.24.0
pydantic>=2.0.0
python-multipart>=0.0.6

# OpenVINO & AI
openvino>=2024.0.0
openvino-genai>=2024.0.0

# Data Processing
numpy>=1.24.0
Pillow>=10.0.0

# HTTP & Networking
requests>=2.31.0
websockets>=12.0  # For WebSocket Realtime API

# Client SDK (for examples/testing)
openai>=1.0.0

# Gradio UI
gradio>=4.0.0

# Document Processing (RAG - optional)
PyPDF2>=3.0.0  # PDF extraction
python-docx>=1.0.0  # Word document extraction

# Audio Processing (TTS/STT)
librosa>=0.10.0  # Audio loading for Whisper
soundfile>=0.12.0  # Audio file I/O
sounddevice>=0.4.6  # Microphone/speaker (voice chat)
pydub>=0.25.0  # Audio format conversion (optional)

# PyTorch (compatible with optimum-intel 1.18.x)
torch>=2.1.0,<2.5.0
torchvision>=0.16.0
torchaudio>=2.1.0

# Model Conversion Tools (for model compilation)
transformers>=4.40.0
accelerate>=0.26.0
huggingface_hub>=0.20.0
sentencepiece>=0.1.99
protobuf>=3.20.0

# Optimum toolchain (vision model support requires optimum-intel >= 1.19.0)
optimum>=1.19.0
optimum-intel[openvino]>=1.19.0
nncf>=2.11.0
hf_xet

# Diffusers (for Stable Diffusion / Text2Image)
diffusers>=0.30.0

# ONNX support
onnx>=1.16.0
onnxruntime>=1.18.0