Skip to content

Commit 4024560

Browse files
aimen-djariaimen-djari
committed
add ai frameworks
1 parent b279ccb commit 4024560

File tree

1 file changed

+138
-0
lines changed
  • src/get-started/protocol

1 file changed

+138
-0
lines changed

‎src/get-started/protocol/ai.md‎

Lines changed: 138 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,138 @@
1+
---
2+
title: iExec for AI
3+
description: How iExec helps develop, deploy and execute confidential AI applications with framework support
4+
---
5+
6+
# 🧠 iExec for AI
7+
8+
The iExec Platform delivers powerful tools specifically for AI developers. Build, deploy, and execute confidential AI applications with enterprise-grade security and decentralized infrastructure.
9+
10+
## 🚀 Quick Start
11+
12+
**Want to get started immediately?**
13+
- 📚 **[AI Frameworks Hello World](https://github.com/iExecBlockchainComputing/ai-frameworks-hello-world)** - Ready-to-use Docker examples for TensorFlow, PyTorch, and more
14+
- 🛠️ **[Build & Deploy](/guides/build-iapp/build-&-deploy)** - General iApp development guide (not AI-specific)
15+
- 🔬 **[TDX App Guide](/guides/build-iapp/advanced/create-your-first-tdx-app)** - Build TDX applications (works well for AI workloads)
16+
17+
## 🛡️ Why iExec for AI?
18+
19+
### Confidential Computing
20+
Your AI models and data are protected end-to-end using Trusted Execution Environments (TEEs):
21+
- **Data Privacy**: AI computations are encrypted throughout processing
22+
- **Secure Training & Inference**: Models and data can never be accessed by unauthorized entities
23+
- **Hardware-Level Security**: Intel SGX and TDX provide military-grade protection
24+
25+
### AI Monetization
26+
Monetize your AI assets easily and securely:
27+
- **Datasets**: Encrypt and sell access to your training data
28+
- **Models**: Deploy and monetize your trained AI models
29+
- **Agents**: Create and sell AI agents and applications
30+
- **Ownership Preserved**: Your digital assets always remain yours
31+
32+
### Decentralized Infrastructure
33+
Scale AI applications without centralized cloud dependencies:
34+
- **On-Demand Compute**: Access powerful resources when you need them
35+
- **Fair Pricing**: Transparent, blockchain-verified execution costs
36+
- **Global Network**: Deploy across a worldwide network of secure workers
37+
38+
## 🤖 AI Framework Support
39+
40+
### Overview
41+
42+
| Framework | TDX Support | SGX Support | Best For |
43+
|-----------|-------------|-------------|----------|
44+
| **TensorFlow** | ✅ Yes (3.01GB) | ❌ No | Deep learning, production ML |
45+
| **PyTorch** | ✅ Yes (6.44GB) | ❌ No | Research, computer vision |
46+
| **Scikit-learn** | ✅ Yes (1.18GB) | ✅ Yes (1.01GB) | Traditional ML, data analysis |
47+
| **OpenVINO** | ✅ Yes (1.82GB) | ❌ No | Computer vision, inference |
48+
| **NumPy** | ✅ Yes (1.25GB) | ✅ Yes (1.08GB) | Scientific computing |
49+
| **Matplotlib** | ✅ Yes (1.25GB) | ✅ Yes (1.08GB) | Data visualization |
50+
51+
### Framework Details
52+
53+
| Framework | Version | Description | TDX Support | SGX Support | Use Cases | Resources |
54+
|-----------|---------|-------------|-------------|-------------|-----------|-----------|
55+
| **TensorFlow** | 2.19.0 | Google's ML framework for production AI | ✅ 3.01GB | ❌ Too large | Deep learning, CV, NLP | [Docs](https://www.tensorflow.org/) • [Quickstart](https://www.tensorflow.org/tutorials/quickstart/beginner) • [Docker](https://github.com/iExecBlockchainComputing/ai-frameworks-hello-world/tree/main/tensorflow) |
56+
| **PyTorch** | 2.7.0+cu126 | Facebook's research-focused DL framework | ✅ 6.44GB | ❌ Too large | Research, DL, CV, NLP | [Docs](https://pytorch.org/docs/) • [Quickstart](https://docs.pytorch.org/tutorials/beginner/basics/quickstart_tutorial.html) • [Docker](https://github.com/iExecBlockchainComputing/ai-frameworks-hello-world/tree/main/pytorch) |
57+
| **Scikit-learn** | 1.6.1 | Comprehensive ML library for Python | ✅ 1.18GB | ✅ 1.01GB | Classification, regression, clustering | [Docs](https://scikit-learn.org/stable/) • [Examples](https://scikit-learn.org/stable/auto_examples/index.html) • [Docker](https://github.com/iExecBlockchainComputing/ai-frameworks-hello-world/tree/main/scikit) |
58+
| **OpenVINO** | 2024.6.0 | Intel's high-performance AI inference toolkit | ✅ 1.82GB | ❌ Execution issues | Computer vision, inference | [Docs](https://docs.openvino.ai/) • [Tutorial](https://docs.openvino.ai/2023.3/notebooks/004-hello-detection-with-output.html) • [Docker](https://github.com/iExecBlockchainComputing/ai-frameworks-hello-world/tree/main/openvino) |
59+
| **NumPy** | 2.0.2 | Fundamental package for scientific computing | ✅ 1.25GB | ✅ 1.08GB | Scientific computing, data analysis | [Docs](https://numpy.org/doc/) • [User Guide](https://numpy.org/doc/stable/user/index.html) • [Docker](https://github.com/iExecBlockchainComputing/ai-frameworks-hello-world/tree/main/numpy) |
60+
| **Matplotlib** | 3.9.4 | Comprehensive library for data visualization | ✅ 1.25GB | ✅ 1.08GB | Data visualization, plotting | [Docs](https://matplotlib.org/) • [Gallery](https://matplotlib.org/stable/gallery/index.html) • [Docker](https://github.com/iExecBlockchainComputing/ai-frameworks-hello-world/tree/main/matplotlib) |
61+
62+
## 🐳 Getting Started with Docker Examples
63+
64+
### What's Included
65+
Our [AI Frameworks Hello World repository](https://github.com/iExecBlockchainComputing/ai-frameworks-hello-world) provides ready-to-use examples:
66+
67+
```
68+
ai-frameworks-hello-world/
69+
├── tensorflow/ # TensorFlow 2.19.0 example
70+
├── pytorch/ # PyTorch 2.7.0+cu126 example
71+
├── scikit/ # Scikit-learn 1.6.1 example
72+
├── openvino/ # OpenVINO 2024.6.0 example
73+
├── numpy/ # NumPy 2.0.2 example
74+
└── matplotlib/ # Matplotlib 3.9.4 example
75+
```
76+
77+
### Quick Start Commands
78+
```bash
79+
# Clone the repository
80+
git clone https://github.com/iExecBlockchainComputing/ai-frameworks-hello-world.git
81+
cd ai-frameworks-hello-world
82+
83+
# Try TensorFlow example
84+
cd tensorflow
85+
docker build -t hello-tensorflow .
86+
docker run --rm hello-tensorflow
87+
88+
# Try PyTorch example
89+
cd ../pytorch
90+
docker build -t hello-pytorch .
91+
docker run --rm hello-pytorch
92+
```
93+
94+
### Features
95+
- **✅ Isolated Testing**: Each framework runs in its own container
96+
- **✅ Reproducible**: Consistent environment across systems
97+
- **✅ TDX Ready**: All containers tested for Intel TDX compatibility
98+
- **✅ Easy Deployment**: Simple build and run commands
99+
100+
## 📊 Technology Comparison
101+
102+
### TDX vs SGX for AI
103+
104+
| Feature | Intel TDX | Intel SGX |
105+
|---------|-----------|-----------|
106+
| **Memory Limit** | Multi-GB+ | ~1.95GB |
107+
| **Framework Support** | All major frameworks | Limited (Scikit-learn, NumPy) |
108+
| **Code Changes** | Minimal ("lift and shift") | Significant modifications required |
109+
| **Production Ready** | ✅ Yes | ⚠️ Limited |
110+
| **AI Workloads** | ✅ Excellent | ❌ Restricted |
111+
112+
### Recommendations
113+
114+
#### For Production AI Applications
115+
- **Use TDX** for TensorFlow, PyTorch, and OpenVINO
116+
- **Use SGX** for lightweight ML with Scikit-learn and NumPy
117+
118+
#### For Development and Testing
119+
- **Start with SGX** for simple ML tasks
120+
- **Migrate to TDX** for complex AI workloads
121+
122+
123+
## 📚 Next Steps
124+
125+
### Learn TEE Technologies
126+
- **[Intel SGX Technology](/get-started/protocol/tee/intel-sgx)** - SGX limitations and capabilities
127+
- **[Intel TDX Technology](/get-started/protocol/tee/intel-tdx)** - TDX advantages for AI
128+
- **[SGX vs TDX Comparison](/get-started/protocol/tee/sgx-vs-tdx)** - Detailed comparison
129+
130+
### Build AI Applications
131+
- **[Build & Deploy](/guides/build-iapp/build-&-deploy)** - Create your first AI application
132+
- **[Build Intel TDX App](/guides/build-iapp/advanced/create-your-first-tdx-app)** - TDX applications for AI workloads
133+
- **[Inputs and Outputs](/guides/build-iapp/inputs-and-outputs)** - Handle data flow in TEE environment
134+
135+
### Explore Examples
136+
- **[AI Frameworks Hello World](https://github.com/iExecBlockchainComputing/ai-frameworks-hello-world)** - Ready-to-use Docker examples
137+
- **[iExec Discord](https://discord.com/invite/pbt9m98wnU)** - Community support
138+
- **[Protocol Documentation](https://protocol.docs.iex.ec)** - Technical deep dive

0 commit comments

Comments
 (0)