Episode
Wan-Streamer v0.2: Higher Resolution, Same Latency
- Podcast
- Daily Paper Cast
- Published
- Jul 8, 2026
- Duration seconds
- 1301
- Processing state
not_requested- Canonical source
- https://share.transistor.fm/s/fcd90bf9
Actions
POST https://stenobird.com/v1/public/podcasts/daily-paper-cast-7079649/episodes/wan-streamer-v0-2-higher-resolution-same-latency/transcription-requests
Idempotently request low-priority transcript generation for this episode.GET https://stenobird.com/podcast/daily-paper-cast-7079649/wan-streamer-v0-2-higher-resolution-same-latency.md
Read the agent-friendly Markdown representation of this episode resource.
Summary
🤗 Upvotes: 26 | cs.CV, cs.AI, cs.GR, cs.LG Authors: Lianghua Huang, Zhi-Fan Wu, Yupeng Shi, Wei Wang, Mengyang Feng, Junjie He, Chen-Wei Xie, Yu Liu, Jingren Zhou, Ang Wang, Bang Zhang, Baole Ai, Chen Liang, Cheng Yu, Chongyang Zhong, Jinwei Qi, Kai Zhu, Pandeng Li, Peng Zhang, Wenyuan Zhang, Xinhua Cheng, Yitong Huang, Yun Zheng, Yuxiang Bao, Yuzheng Wang, Zoubin Bi Title: Wan-Streamer v0.2: Higher Resolution, Same Latency Arxiv: http://arxiv.org/abs/2607.04443v2 Abstract: We present Wan-Streamer v0.2, a latency-preserving upgrade of the native-streaming, end-to-end audio-visual interaction model. v0.2 keeps the v0.1 modeling formulation, but raises the interactive output stream from 192x336 to 640x368 while preserving approximately 200 ms model-side signal-to-signal latency at 25 FPS. The higher-resolution stream supports scene-grounded mid-shot agents whose posture, gaze, hands, nearby objects, and local scene layout remain legible during real-time conversation. To support the larger visual stream without adding user-visible delay, v0.2 keeps the thinker as a single-GPU low-latency path for streaming perception, the short language/state Transformer pass that builds the generation cache, and final decoding. The performer becomes a multi-GPU Ulysses-style context-parallel group for the expensive next-unit latent generation. Each performer rank writes incoming K/V into a pre-sharded local cache. The long high-resolution latent video sequence is split across ranks for denoising and gathered through Ulysses communication, while the much shorter audio latent sequence is generated without sequence sharding. In this split, the thinker's language/state computation reaches the performer only as K/V conditioning, so no separate language sequence has to be communicated inside t…