X just quietly rolled out a toggle in its post composer that lets users stop xAI's Grok from generating alternate versions of their uploaded media. The move follows regulatory probes and massive looming fines after Grok was abused to generate non-consensual deepfakes in early January. Why it matters: Giving users an opt-out control helps protect personal content while potentially saving X millions in penalties.
Researchers just open-sourced CubeComposer, a video model that converts standard single-camera video into full 360-degree panoramic footage with 4K enhancement. By breaking frames into six cube faces and using sparse attention, it fills in what was behind the lens far better than tools like Argus, though minor distortion remains.
Tencent researchers just introduced EffectMaker, a framework that clones visual effects from a reference video onto new subjects. It pairs a multimodal LLM with a diffusion transformer to adapt complex effects without per-effect training, backed by a massive 130k-video dataset.
Tencent just open-sourced HY-WU, a powerful image editing framework designed for high-fidelity virtual try-on and style transfer. It generates on-the-fly LoRA adapters per prompt to dress subjects in reference outfits without test-time fine-tuning, nearly matching Nano Banana. The only hurdle is compute: running its base model currently requires at least four 80GB GPUs.
RealWonder just open-sourced a system that lets users apply virtual forces to a single image and watch objects move in real time. Powered by a physics simulation bridge and a 4-step diffusion generator, it hits 13.2 FPS at 480p on an H200 GPU. It brings structural physical understanding to AI video, with weights and code fully public.
Luma just introduced Luma Agents, a creative agent platform powered by its new Uni-1 unified multimodal model. The system handles end-to-end ad campaigns across text, images, video, and audio, using an iterative self-critique loop while orchestrating third-party models like Google's Veo 3 and ElevenLabs. It is now rolling out gradually via API.
Researchers just introduced OmniLottie, a framework that leverages vision-language models to generate lightweight Lottie JSON vector animations directly from text, image, or video prompts. Alongside the model, the team introduced MMLottie-2M, a benchmark dataset of two million annotated vector animations, making automated UI motion design far more practical.
ByteDance and Stanford researchers just open-sourced Spectrum, a training-free framework that accelerates image and video diffusion models by up to 4.7x. By forecasting future step trajectories with Chebyshev polynomials, it lets models like FLUX.1 and Hunyuan Video generate outputs much faster with virtually no drop in quality.
Researchers just open-sourced WildActor, a video generation framework that keeps an identity's face, body, and clothing consistent across unconstrained motions using only three reference images. Built on Wan2.2-5B, it supports both real people and cartoon characters while releasing the Actor-18M data construction pipeline.
Researchers just introduced ArtiFixer, a generative framework designed to repair sparse 3D reconstructions. By pairing an opacity-mixing model with an auto-regressive diffusion pipeline, it hallucinates missing backgrounds and cleans up artifacts when inputs are limited to a handful of photos, beating previous baselines by up to 3 dB PSNR.
ByteDance researchers introduced a spatial reward modeling framework alongside SpatialScore, fixing a stubborn flaw where image generators mix up object positions. By training on over 80,000 spatial preference pairs, the method helps models like Flux follow complex layout prompts. The team plans to open-source the dataset and weights soon.
SynapsCAD just launched an open-source tool aiming to become the Cursor for CAD. Developers write OpenSCAD code on the left and see real-time 3D geometry on the right, with the ability to click any part in the viewport to have AI modify the underlying code. Why it matters: Unlike black-box generative 3D meshes, every AI change remains clean, version-controlled code.
Google just launched Nano Banana 2 (Gemini 3.1 Flash Image), blending the reasoning and fidelity of its Pro model with blazing-fast Flash generation. It handles subject consistency for up to five characters, real-time web search grounding, and outputs ranging from 512px to 4K resolution. Why it matters: High visual consistency and real-time facts are usually slow and expensive; bringing both to a fast-tier model makes rapid storyboarding and production design vastly more accessible.
QuiverAI just opened public beta access to Arrow 1.0, a generative model built specifically for creating SVG vector graphics rather than standard raster images. It directly outputs scalable, editable vector assets, saving designers and developers the hassle of manual vector tracing.
Google just redesigned its creative platform Flow, folding experimental tools ImageFX and Whisk directly into a single workspace. Creators can now generate assets with the built-in Nano Banana model and immediately use them as ingredients for Veo video generations, alongside new lasso-based natural language precision editing.
PhysicEdit just open-sourced a physics-aware image editing framework that actually understands real-world causal dynamics. Built on top of Qwen ImageEdit, it reliably simulates complex visual physical phenomena like light refraction in water, melting butter, and collapsing card houses where most flagship models fail.
Researchers just introduced VecGlypher, an open multimodal model that generates complete vector glyphs directly from text prompts or reference images. Instead of converting blurry raster graphics to vector paths, it autoregressively emits clean SVG outlines in one pass. It noticeably outperforms general models like GPT and Gemini in font stability, offering designers a powerful new foundation for custom typography.
Researchers just open-sourced tttLRM, a novel 3D reconstruction model that uses Test-Time Training to convert multiple photos into crisp 3D Gaussian Splats. Unlike traditional 3DGS, it strips away foggy artifacts to capture fine wires and tiny text, all within a compact sub-4GB model that runs on consumer GPUs.
Researchers just introduced SARAH, a real-time conversational agent for VR that actually moves like a human. It tracks your 3D head position to turn toward you, adapt its posture, and hold eye contact at over 300 FPS. Users can even dial in gaze intensity, making VR avatars feel far less like rigid digital puppets.
Researchers just introduced a human-centric video world model that turns tracked head and hand poses from a VR headset into real-time, egocentric interactive video. Running at around 11 FPS, it lets users physically interact with generated environments instead of relying solely on text or keyboard controls. While fidelity is still early, it marks a significant step toward real-time generated XR.
