Video Editing

Magnific published a workflow that creates a depth map from reference footage with Claude Opus 5.5 and its MCP, then uses Seedance 2.5 to alter the environment, wardrobe, or characters. The edits retain the source motion while changing those visual elements.

Creator demos show Claude Opus 5.5 producing short narrated explainers, opening credits, and marketing videos from detailed prompts. One report says a 30-second voiced animation ran inside Claude's VM from a single prompt.

Creators describe using InVideo Editor with GPT-6 Astra to turn raw footage into edits while retaining control over story and style. In one test, the editor added atmosphere, impacts, transitions, and synchronized effects to the edit.

A creator’s tool uploads video and exposes frames through volumetric playback, time slices, opacity controls, synchronized playback, and 360-degree orbit export.

Charas Power AI documented a workflow using InVideo Agent Two to produce a nearly four-minute fantasy trailer from script development through shot generation. The workflow locks character, environment, and prop sheets before generation.

Higgsfield’s demo shows GPT-6 Astra handling camera work, 3D rendering, motion blur, grain matching, and compositing for a live-action VFX shot. A separate demo claims Astra retains context in After Effects.

Higgsfield says its ChatGPT integration turns uploaded assets, product sites, or briefs into editable After Effects work. Higgsfield also describes automatic reframing and a prompt-built GPU plugin for animating a 3D brain.

Black Forest Labs launched FLUX 3 Video Edit [fast] for targeted edits to objects, backgrounds, text, materials, events, and dialogue. The company says the mode better preserves unedited parts of a clip while making the specified change.

Netflix released ID-V2V to restyle recorded footage while preserving a performer’s identity, expression, gaze, motion, and lip sync. The Apache 2.0 release supports relighting and imperfect edited keyframes.

A creator used InVideo Agent Two to make a nearly four-minute trailer using a script, character and location sheets, visual references, and a project playbook. The workflow iterated Seedance 2.5 shots before final editing.

Higgsfield says GPT-6 Astra can clean footage in After Effects, assemble sound effects in Logic Pro, and deliver an editable project file. Separate demos show it generating, editing, and posting short videos through Higgsfield and TikTok.

Topaz for Web brings video enhancement to a browser app, including upscaling, slow motion, frame interpolation, and SDR-to-HDR conversion. It also includes image upscaling and deblurring.

Google’s Agentic Video processing lets Gemini reason across a video’s transcript, audio, and frames rather than scan the whole file uniformly. Google says dynamic frame sampling can cut token use by up to 88% on supported tasks.

A creator used InVideo’s Agent Two to plan a nearly four-minute film, then moved generated and uploaded footage into an editable timeline. The editor agent assembles footage from natural-language instructions while preserving manual control.

Magnific used its agent to plan scenes and prompts for a character-led romantic trailer. It assembled the MiniMax H3 clips in Premiere through its plugin.

Runway released Ruby to convert uploaded or generated SDR clips into 16-bit HDR. Max and Enterprise users can export clips up to 30 seconds as EXR sequences, ProRes, or HEVC files.

Topaz says Hyperion 2.5 converts 8-bit AI footage into HDR video. Astra now exports ProRes 10-bit, EXR 16-bit, and H.265 10-bit; Topaz Video support is planned.

Remotion Skills 2.0 simplifies its APIs, generates interactively editable code in Studio, routes best-practice prompts to relevant skills, and removes embedded design defaults. Studio output remains editable, while projects no longer receive built-in design defaults.

Stages says V5 gives VIDX and SIGNAL shared editor intelligence, durable command history, deterministic CUE actions, and safe replay. Stages Connect can control TouchDesigner, Blender, After Effects, and Unity through MCP.

Luma says Layers is live in Luma Agents for extracting editable elements from uploaded or generated images. Demos show product swaps, headline edits, object inserts, and modify-video keyframes.

Adobe's AI Assistant is being demoed as a storyboard-to-video pipeline inside Firefly. The flow creates character sheets, shot lists, frames, soundtrack, clips, and a rendered board; Firefly Boards also has free daily generations.

Creators posted concrete pipeline notes for Stages AI, LTX, inVideo, Seedance, Firefly, and Opus 5. Reports included a $12 avatar tutorial, two-day agent runs, and QC loops for short films.

Stages AI demoed VIDX color grading with AI, traditional editing, adjustment layers, and Aleph 2 usage through Stages credits instead of direct Runway accounts. Dustin Hollywood also teased STAGES PRO v5 creator workflows and presets.

Venturetwins built a podcast clipping app with Thinking Machines’ Inkling. The app analyzes long-form audio, selects clip candidates by topic or best moment, and directs FFmpeg edits.

LLMJunky said Codex used Palmier with FFmpeg to choose frame ranges, speed up a clip, add overlays, animate an icon, and place text. He still handled some zoom effects manually.

Creators report Gemini Omni Flash is now available inside Adobe Firefly Boards. Early tests focused on video restoration and restyling prompts, with broader mobile rollout mentioned in replies.

Kaigani posted generated-video QC tests where Gemini caught subtle errors and said they are deliberately glitching clips for a benchmark. The work follows earlier Codex frame-analysis tests and focuses on detecting defects in generated video.

Figma Motion demos showed designers prompting animations while retaining manual parameter control and auto-keyframes for shader and button effects. Try it for non-destructive animation edits across videos, frames, and objects.

Magnific released a Photoshop plugin for generation, upscaling, retouching, background removal, relighting, reframing, feed browsing, and stock search. Try it for image-editing workflows; same-day demos showed Gemini Omni motion graphics and Seedance 2.0 character-sheet consistency inside Magnific.

CapCut opened its CRE[AI]TE festival with $200,000 in prizes across film, series, creative, and commercial tracks, and winners are slated for a major festival screening. Submit with CapCut Video Studio if you want eligibility for the funding and the exposure tied to that workflow.

Magnific rolled Gemini Omni Flash into its MCP and Spaces workflow with claims that edits keep the full scene in memory across turns, while Figma Weave also added access in its Google-model bundle. That expands Google’s video-editing model beyond AI Studio into creator tools built for iterative asset work.

Google launched Gemini Omni Flash for video generation and editing in the Gemini API and AI Studio at $0.10 per second. Runway, Magnific and Higgsfield rolled it out on day one, so creators can test VFX, real-footage edits and voice-directed scene changes.

Multiple creator demos and launch posts show OpenArt Director generating one-minute cinematic sequences through chat while preserving a timeline for shot, audio, and subtitle edits. That matters because it pushes OpenArt past one-off clip prompting into repeatable finishing workflows with stronger character and style continuity.

Creator testing around Figma Motion now shows shader prompts, canvas handles, and keyframe retiming in the beta, alongside a detailed tutorial. Designers can keep more motion polish inside Figma before handing work to separate animation or code tools.

Remotion released Effects with more than 50 built-in shader transitions and post-processing tools that can be edited interactively, programmatically, or through agents. The release also adds keyframing and component support, including HtmlInCanvas, for richer motion workflows.

Creators are using Gemini Omni to read a reference design and generate a final prompt for another video model while preserving face, voice, lip sync, and gestures. Use it to separate style translation from generation, but plan around the current 10-second output limit.

DrSadek published a full AgentOne recipe that starts with one 6-panel GPT Image 2 storyboard, animates each beat in Seedance 2, then finishes in DaVinci Resolve with a Suno score. Follow the workflow to see where reverse-gravity prompts fail and how the time-reverse edit recovers the shot.

A new creator comparison split Astra 2 into AI-artifact repair and re-detailing, while Starlight Precise 2.5 handled sharpening and recovery on cleaner or real footage. That matters because it gives a concrete two-pass order instead of treating both Topaz models as interchangeable upscalers.

Luma added Timeline with EDL Export, letting editors assemble cuts on full-resolution files and send frame-accurate handoffs into finishing tools. It matters because the platform is moving from clip generation toward a fuller production stack a day after launching Skills.

David Holz said he shot and edited the Midjourney Medical launch video himself and built its realtime browser visualization with three.js/WebGL, Claude, and Codex. That turns the hardware reveal into a creator-side production case study, even if the process is still a one-off launch build.

New creator demos show InVideo Agent One building trailers and shorts through iterative direction inside one workflow. The examples extend the product from cross-device editing into story, voice, scene, and rough-cut collaboration.

Pika introduced Director’s Suite as an agentic workspace for concepting, casting, storyboards, generation, and edits in one project. Creator demos show the same stack assembling a 6-minute pilot in one interface, pushing AI video toward episode-length production.

STAGES AI said Resolve Audio Studio v1.01 shipped for paid users after earlier preview posts, with multi-project sound editing as the main change. Posts also kept v1.02 stem separation on the roadmap, moving the audio module from preview into live use.

NVIDIA released SANA-Streaming as a system for real-time editing of minute-long video, including scene swaps, style transfer, inpainting, and robotics overlays. The demo emphasized temporal consistency across longer clips rather than one-off image transformations.

Runway opened a ChatGPT integration so users can generate and edit Runway video and image outputs without leaving a chat thread. Creator tests around Aleph 2.0 also found stronger reference-based control, even when Seedance looked more natural in side-by-side results.

Pixlo added new transition presets including push, zoom, dissolve, rise, and drop for still-image recap videos. Use it if you want to keep the workflow at image upload plus export instead of moving into a separate editor.

Curious Refuge published a four-stage restoration workflow that runs old footage through Seedance 2.0, GPT Image 2 still remastering, style rematching, and a final Topaz Astra pass. Apply it as a repeatable pipeline for archive cleanup instead of relying on a one-shot prompt.

Google says Gemini Omni Flash now leads its benchmark set across image-to-video, text-to-video, and video editing, with API access coming soon. The claim matters because creators are already showing Flow-based reconstructions and relighting demos, but the broader developer rollout is not live yet.

CapCut opened its first global AI festival with Film, Series, Creative, and Commercial tracks plus $200,000 in prizes. The event matters because AI-assisted work now has a cash-backed festival path tied to CapCut Video Studio, so creators can submit outside isolated prompt contests.

Pika expanded its Language Swap story with official multi-language demos and creator tests of own-voice Japanese dubbing through MCP. The evidence matters because localization is shifting from voice replacement toward performance-preserving video translation inside creator workflows.