Privacy-preserving LLM inference with CKKS homomorphic encryption and Private Linear Layer (PLL) protection for LoRA fine-tuned models
-
Updated
Oct 18, 2025 - Python
#
secure-inference
Here are 6 public repositories matching this topic...
Scoria AI is a decentralized AI agent framework on blockchain, enabling private, on-device Web3 intelligence for users and enterprises.
agent rust typescript ai cuda oracle dao web3 tensorrt onnx federated-learning on-chain solana zero-knowledge-proofs defi secure-inference zkml
-
Updated
May 9, 2025 - Rust
Scoria AI is a decentralized AI agent framework on blockchain, enabling private, on-device Web3 intelligence for users and enterprises.
agent typescript cuda dao web3 tensorrt onnx federated-learning on-chain solana zero-knowledge-proofs defi secure-inference zkml
-
Updated
Jan 20, 2026 - Rust
Isolated runtime for Hugging Face models requiring `trust_remote_code=True`, using a loopback-only Docker container and a FastAPI bridge.
python docker machine-learning sandboxing model-server ai-security fastapi huggingface secure-inference llm trust-remote-code
-
Updated
Dec 20, 2025 - Python
GPT-OSS B20 Local Execution. Lightweight local environment for running it with Python 3.12 and CUDA acceleration. - Run GPT-OSS B20 entirely offline - Optimize text generation with GPU - Enable fast, secure inference on consumer hardware.
text-generation performance-optimization gpu-optimization edge-ai inference-acceleration secure-inference model-runtime minimal-setup llm-inference open-source-llm local-execution offline-inference privacy-preserving-ai consumer-gpu gpt-oss-b20 lightweight-environment
-
Updated
Aug 13, 2025 - Python
SPRINT is a scalable framework for differentially private (DP) fine-tuning and inference via multiparty computation (MPC) of transformer-based models. SPRINT is built on top of PyTorch, Opacus for DP fine-tuning and CrypTen for MPC inference.
-
Updated
Oct 30, 2025 - Python
Improve this page
Add a description, image, and links to the secure-inference topic page so that developers can more easily learn about it.
Add this topic to your repo
To associate your repository with the secure-inference topic, visit your repo's landing page and select "manage topics."