Wan

Timeline

1. Oct 8, 2026 · Research paper · 1 source
   WorldAlign: Decoupled 4D Reward for World-Consistent Video Generation
   Faithful visual world simulation requires generated videos to maintain 4D world consistency, encompassing both static and dynamic consistency.
2. Oct 8, 2026 · Open-source release · 1 source
   unslothai/unsloth v0.1.905-beta: Sandboxing is here!
   We're introducing Windows, Mac and Linux sandboxing in Unsloth!
3. Oct 8, 2026 · Research paper · 1 source
   Memory Forcing: Attendable Mid-Horizon History for Streaming Video Generation
   Autoregressive video diffusion enables causal video streaming without a bidirectional pass over the full clip, but existing few-step systems usually retain only the opening and most recent frames in a fixed-size KV cache.
4. Oct 8, 2026 · Research paper · 1 source
   Towards Unified Evaluation of Prompt Enhancers for Video Generation
   To address this gap, we introduce PEBench, the first unified benchmark for direct PE evaluation across text-to-video, image-to-video, and reference-to-video prompt enhancement.
5. Oct 8, 2026 · Research paper · 1 source
   Parametric Trajectory Distillation for Few-Step Video Generation
   We introduce Parametric Trajectory Distillation (PTD), which lets the student parameterize teacher trajectory segments as polynomials and learn from teacher guidance along its own predicted path.
6. Oct 8, 2026 · Research paper · 1 source
   iCATS: Fast Video Generation via Interaction-Aware Sparse Attention and Timestep-Adaptive Sparsity
   Training-free sparse attention offers a practical acceleration solution to Diffusion Transformers (DiTs) via reducing computations without fine-tuning.
7. Oct 7, 2026 · Research paper · 1 source
   Fluid-Gen-Zero: Grounding Pretrained Video Generators in Physics without Training
   We present Fluid-Gen-Zero, a training-free framework for physics-aware fluid-object interaction video generation that decouples physical reasoning from appearance synthesis.
8. Oct 7, 2026 · Research paper · 1 source
   GRACE: Generation-aware latent compression for efficient video generation
   To address this, we propose Generation-Aware Latent Compression for Efficient Video Generation (GRACE), a two-stage framework that compresses a pretrained video autoencoder while keeping it compatible with the pretrained DiT.
9. Oct 7, 2026 · Open-source release · 1 source
   unslothai/unsloth v0.1.904-beta: Train your own Decision model
   Turn any text or vision LLM into a Jev-style decision model in Unsloth, with decision accuracy going from 30% to 80%.
10. Oct 6, 2026 · Research paper · 1 source
    Backend-Agnostic Sparse Attention for Fast High-Resolution Visual Generation
    To tackle these challenges, we propose BASA, a backend-agnostic sparse attention, which brings the best of both worlds: visual quality and practical acceleration.
11. Oct 6, 2026 · Research paper · 1 source
    OpenWAM: An Open Framework for Composable World-Action Models
    We introduce OPENWAM, an open world-action modeling framework built around a common causal robot-video foundation and configurable video-action interaction.
12. Oct 6, 2026 · Open-source release · 1 source
    huggingface/diffusers v0.41.0: Diffusers 0.41.0: QwenImage 2.1 pipeline and more
    > This release brings Qwen-Image 2.1 to Diffusers, with text-to-image generation, image editing, native transparency, and LoRA training.
13. Jun 13, 2026 · Open-source release · 1 source
    sgl-project/sglang v0.5.13
    DeepSeek V4 — context parallelism & sparse-attention kernels: Building on the v0.5.12 Day-0 path, v0.5.13 extends DeepSeek-V4 to context-parallel serving and adds its sparse-attention kernels:

Often appears with