CMC research separates video motion from reference appearance
A new preprint studies motion transfer while limiting copied appearance.
Generative Media
Generative media covers the AI systems used to create and edit images, video, speech and music. ByteForward follows creative model releases and product updates with attention to what creators can control and where results remain unpredictable. Topics include text to image generation, AI video tools, voice synthesis and the workflows that connect them. Output quality is only part of the picture when a project also depends on consistency, editing flexibility and generation time. Access limits, pricing and licensing terms help determine whether a tool fits a particular creative use. Our AI model release coverage tracks new systems and the practical details of their rollout. Explore AI culture and society for reporting on how generated media affects creative work, audiences and trust. Browse the stories below to understand both the possibilities and the constraints behind the latest generative AI media tools.
MTOR study combines captions and temporal patterns to detect AI video
Researchers test synthetic video detection across 46 generator variants. Code and model releases remain pending.
Google releases Nano Banana 2.1 in the Gemini API
Nano Banana 2.1 is available in the Gemini API. Google’s current documentation lists no shutdown date for the older Nano Banana 2 model.
BioRender introduces Leo for planning editable scientific figures
BioRender’s Leo guides figure planning with sketches and editable drafts. Access requires a paid add on plan.
Suno adds album publishing with follower notifications
Creators can group songs into an album or convert an existing playlist into a release.
OpenAI to test visual ads during ChatGPT image generation
A US test is planned for later October as OpenAI expands conversion tracking and brand suitability work.
Claude Opus 5.5 motion graphics show why the setup matters
Creator examples span quick product explainers, long rendering jobs and editable animal models. The brief and tools shape the result.
Luma turns one static ad into versions for different placements
Choose the sizes and languages, generate a batch, then check each version against the approved design.
Audible adds cast guides and plans AI character conversations
Audible adds cast guides and plans interactive extras for a limited beta
InvokeAI 7 alpha organizes creative work into projects
The experimental update brings a redesigned workspace and audio support
Lightroom adds Prompt to Edit with Nano Banana 2
Lightroom desktop 9.6 brings early access editing from text prompts, separate DNG outputs and generative credit costs
ASCI finalizes synthetic advertising guidance
India’s advertising standards body sets disclosure tests for synthetic media and sponsored AI recommendations
QwenCloud adds Qwen Image 2.1 Pro API access
The October 2 hosted API addition supports transparent images and editing with up to ten references.
Tavus previews Griffin for live AI video conversations
Tavus is testing Griffin Lite with selected research participants. Its video conversation results come with important limits on access and evaluation.
HeyGen Video brings scene generation and audio to its API
HeyGen Video generates complete scenes with sound from prompts and references. October launch rates depend on resolution and whether video references are used.
Runway Ads links AI creative generation with campaign results
Runway Ads is piloting a system for generating advertising creative, publishing approved variants and using campaign results to guide the next batch.
Suno opens Speech beta for narration with generated music
Suno Speech creates spoken narration and background music in one track. The public beta is available on web and mobile, with uneven accents and timing still possible.
FLUX 3 Image launches with layout controls and 4K output
Black Forest Labs opens a dedicated FLUX 3 image endpoint with layout controls, reference editing and output up to 4K. Pricing and preservation limits matter.
Alibaba opens Wan 3.0 video generation to public beta testers
While the rest of the week argued about coding agents, Alibaba opened a public beta for Wan 3.0. The product page is selling longer clips, more reference assets, and a free-quota on-ramp that turns into a per-second bill.
xAI ships Grok Voice Think Fast 2.0 — time-to-first-audio hits 0.70s
xAI’s Grok Voice Think Fast 2.0 ships speech-to-speech at 0.70s TTFA and $0.08/min, becoming the latest grok-voice alias for low-latency voice agents.
Google brings Lyria 3.5 music generation to free Flow Music
DeepMind’s Lyria 3.5 music model lands in free Flow Music with richer melodies and longer songs, pushing AI music further into everyday listening habits.
Google Vids adds Gemini Omni editing and personal avatars
Google just made Workspace video feel less like a slide deck with motion and more like a generative editor. In a Workspace blog post, the company says Gemini Omni and personal avatars are rolling into Google Vids, so teams can…