{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreiacfrmx4qdojeykd22b5y2cj4ozhkzayx7izycpnukb6dpp2kufsy",
    "uri": "at://did:plc:4jjxx3max7tcdxwmdkjrnyj4/app.bsky.feed.post/3mozvaduwphc2"
  },
  "coverImage": {
    "$type": "blob",
    "ref": {
      "$link": "bafkreifd3aqxycwvhadrsbzmuofqmjsifmfyhyghrq33zvjx4iks7xza5m"
    },
    "mimeType": "image/webp",
    "size": 27468
  },
  "path": "/seedance-2-5-updates",
  "publishedAt": "2026-06-23T20:23:40.000Z",
  "site": "https://nofilmschool.com",
  "tags": [
    "Bytedance",
    "Seedance 2.0",
    "Seedance 2.5",
    "Generative ai",
    "Ai video",
    "Ai video model",
    "Seedance updates",
    "A24 is partnering with Google",
    "ByteDance’s Seedance 2.0"
  ],
  "textContent": "\n\n\n\nIn the ever-changing landscape of AI, it feels like major updates and breakthroughs often come in waves. We’ll have a few weeks or months of downtime where everything returns to normal, and the AI companies and their fans kind of mind their own business, then bam—we get a bunch of announcements all at once.\n\nFollowing news that A24 is partnering with Google to develop AI-powered tools for filmmakers, we now hear that ByteDance’s Seedance 2.0 is getting an update, and that a v2.5 model is on the way.\n\nIf you’re morbidly or regularly curious about this news, here’s what you need to know about Seedance 2.0 and a likely Seedance 2.5 coming soon.\n\n* * *\n\n### Seedance Updates on the Way\n\n\n\n\n> So, we’re saying it sounds like it’s kind of hard to find official sources for any of this news, but it is spreading like wildfire on the internet (mostly on X) that Seedance has some updates on the way. The first bit of information concerns the current version of Seedance (2.0), which is reportedly now able to support native 4K with 10-bit color depth.\n\nByteDance is also set to introduce a new Doubao 2.1 Pro language model, a Seedream 5.0 Pro image model, and its new Seed-Audio 1.0 audio model, all major AI models set to challenge the biggest and most popular players in the space.\n\nThe other big headline being shared online is that a Seedance 2.5 generative AI video model update is on the way and will reportedly also be able to generate clips up to 30 seconds long, something that’s been rare for AI video models to do without cross-stitching generations.\n\n### What Comes Next?\n\n\n\n\n> As is the case with any news about Seedance, though, it’s hard to tell what is fact and what's fiction. And even when facts do come out, the models can be hard to find and use, as many avenues require users to create accounts with other AI platforms like Higgsfield or CapCut to access them.\n\nThe same reports online all seem to share the same sentiment that a Seedance 2.5 model is set to launch here in early July. However, no word on whether or when the model will be added to the same AI video platforms for easier access.\n\nIn many ways, we’re happy to report, the generative AI video buzz has kind of slowed down as of late, as, besides a few projects here and there, we really haven’t seen much widespread adoption of this form of content creation.\n\nWill that change in the future? Likely yes to a degree. Will that change come sooner with a Seedance 2.5? Only time will tell. We’ll keep you posted, though!",
  "title": "Looks Like a Seedance 2.5 Model is on the Way With 30-Second Generation Capabilities"
}