LServe accelerates long-sequence LLM serving with unified sparse attention for both prefilling and decoding, achieving up to 3.3× speedup over state-of-the-art solution without sacrificing accuracy.

QServe accelerates large-scale LLM serving on GPUs with QoQ (W4A8KV4) quantization, boosting the generation throughputs by up to 3x over the state-of-the-art solution.

A new W4A4 quantization paradigm for diffusion models.

We propose COAT, a memory efficient FP8 training method for large language models.

SANA is an efficient linear DiT that can generate images up to 4096 × 4096. SANA delivers: 20x smaller & 100x faster than FLUX; Deployable on laptop GPU; Top-notch GenEval & DPGBench results.

A new family of high-spatial compression autoencoders for accelerating high-resolution diffusion models.