intel/neural-compressor helps research and automate market workflows

Описание и применение

intel/neural-compressor: open-source проект в категориях Trading, AI/ML, Database. Практическое применение: research trading systems; backtest strategies. Уровень готовности: готов для серьезного pilot с проверкой license, security и maintainer activity.

Задача

intel/neural-compressor helps research and automate market workflows.

Результат

High for an OSS candidate: recent activity, adoption, and ecosystem signals are strong enough for a serious pilot.

Рабочий процесс

• research trading systems
• backtest strategies
• monitor market workflows

Доказательства

SOTA low-bit LLM quantization (INT8/FP8/MXFP8/INT4/MXFP4/NVFP4) & sparsity; leading model compression techniques on PyTorch, TensorFlow, and ONNX Runtime

Источник