← Back to all sparks
P

Picsart

DESIGN
Velocity8.8

AI-powered photo and video editing platform with creative design tools

Picsart bolts on every new video model within days — WAN 3.0 pushes single takes to 30 seconds.

ai-modelsvideo-generationmodel-aggregationagentic-accesscreator-toolscontent-marketing
Current state
Picsart's feed runs at high volume and is mostly comparison and how-to marketing content. The genuine product news is a narrow band inside it, identifiable by whether the headline names Picsart itself: WAN 3.0 arriving with 30-second generation, and a Picsart MCP endpoint with a Motion Studio skill that lets an assistant cut long video into vertical formats. Everything else in the last week is Ideogram-versus-Flux positioning and trend-drop tutorials.
Where it's heading
The integration cadence is the strategy. Third-party image and video models land in Picsart within days of their public release, and the surrounding posts argue that the models themselves have converged — same resolution, same text rendering, same structured prompting — which leaves the aggregation layer and the workflow tools as the actual product. Video is where the additions concentrate: 15-second clips became 30-second single takes across both Seedance 2.5 and now WAN 3.0.
Prediction
Expect the next publicly released image or video model to appear in Picsart within a week, paired with a comparison post, and expect the agent-facing surface to widen beyond the single Motion Studio skill.

Recent moves

  1. 1d ago

    Ideogram 4 vs Flux 2: layout control or consistency

    A comparison post arguing that Ideogram 4 and Flux 2 have converged on resolution, text rendering and structured prompting. It is positioning content for the model catalog, not a change to Picsart itself.

    View source ↗
  2. 1d ago

    How to become an AI artist and earn money with your artwork

    Career-advice content aimed at pulling AI-curious creators into the funnel. No product surface changes.

    View source ↗
  3. 1d ago

    How to make a flying dragon video while preserving character consistency

    A tutorial on holding character consistency across a 30-second video, riding the longer-take capability the recent model additions unlocked. Instructional, not a release.

    View source ↗
  4. 1d ago

    How to use AI motion control for character animation

    An explainer on motion-transfer from a reference clip onto a still portrait. It describes the technique generically rather than announcing a Picsart capability.

    View source ↗
  5. 1d ago

    WAN 3.0 is coming to Picsart with 30-second video

    WAN 3.0 joins the catalog with unbroken 30-second clips and document or web page as a source input, up from 15 seconds and text-only. It extends the same aggregation pattern as Seedance and Flux rather than changing what Picsart is, but it moves the ceiling on clip length again.

    View source ↗
  6. 1d ago

    Daily Trend Drop Vol. 85: “Zoom Out Reveal” – You, Then the Whole City

    Another entry in the daily trend series, packaging a social video format as a template. Recurring editorial filler around the model catalog.

    View source ↗