Generates 480p video with two-step autoregressive diffusion, reaching about 45 ms per frame on eight B200 GPUs. Distilled from Wan2.1-14B, it targets interactive generation but needs a multi-GPU setup for that speed.
Handles text and image inputs and generates text, with a context window of up to 1 million tokens. Its hybrid sparse and sliding-window attention supports long-context work, while the MoE design activates about 12B parameters per token.
Provides six open language models from 0.9B to 375B-A23B for reasoning and agentic tasks. The paper also introduces sparse attention and a decoding speedup that needs no separate draft model.
Details the Xing 4 model architecture, Ascend-scale pretraining systems, and agentic post-training, including a 285B-parameter training run on 8,192 Ascend 910C cards.
Reuses decoder-layer weights to increase computation depth at roughly 9B parameters, taking text and images as input and generating text with up to 512K-token context. It targets reasoning, coding, and tool use, though performance on complex long-horizon tasks can be inconsistent.
Provides a web chat and app front end for Alibaba's Qwen model family, with open-weight language, coding, vision, audio, image, and reasoning models. Its appeal is breadth; its tradeoffs are policy constraints and shifting model availability.