← all releases
Luce-Org·May 1, 2026·v1.0·models
Luce PFlash
Luce PFlash is a C++/CUDA speculative prefill system that achieves ~10.4x faster Time-To-First-Token on long-context inference. A small Qwen3-0.6B drafter scores token importance so the heavy 27B target only prefills the spans that matter — cutting 128K token prefill from 257s to 24.8s on a single RTX 3090.
InferenceCUDALLMOptimizationOpen SourceLocal AI
overview
Luce PFlash is a C++/CUDA speculative prefill system that achieves ~10.4x faster Time-To-First-Token on long-context inference. A small Qwen3-0.6B drafter scores token importance so the heavy 27B target only prefills the spans that matter — cutting 128K token prefill from 257s to 24.8s on a single RTX 3090.
key features
- 01~10.4x faster TTFT on 128K context vs standard llama.cpp
- 02Small 0.6B drafter scores token importance to skip irrelevant spans
- 03Single daemon compress command integrates into existing dflash stack
related productsbrowse all →
Protenixopen sourceProtenix is an open-source, trainable PyTorch implementation of AlphaFold 3, designed for high-accuracy biomolecular structure prediction. It aims to advance accessible and extensible research tools for the computational biology community. ([github.com](https://github.com/bytedance/Protenix?utm_source=openai))Z-Imageopen sourceZ-Image is an efficient image generation foundation model developed by Tongyi-MAI, designed to produce high-quality, diverse, and stylistically versatile images. It serves as a robust base for creators, researchers, and developers seeking advanced image generation capabilities.