# Seedance 2.5 Flash: Release Date and API Access > Check Seedance 2.5 Flash’s announced rollout, API access and confirmed limits, with the differences from standard Seedance 2.5 before you switch. - Canonical HTML: https://omniakey.com/blog/seedance-2-5-flash - Markdown: https://omniakey.com/blog/seedance-2-5-flash.md - Published: 2026-11-12 (updated 2026-11-12) - Author: Kiki, COO at OmniaKey - Language: en - Category: Guide - Reading time: 6 min - All posts: https://omniakey.com/blog/llms.txt --- BytePlus schedules online experience and API calls for Seedance 2.5 Flash from 14:00 UTC+8 on November 15, 2026, according to the availability notice on [Create a video generation task API](https://docs.byteplus.com/en/docs/modelark/create-video-generation-task-api). That notice was checked on November 12, 2026 in New York (November 13 UTC). Standard Seedance 2.5 is a separate model with a 1080p option and a longer published feature list, and work that needs 1080p or those capabilities stays on the standard model. - **Opening:** November 15, 2026 at 14:00 UTC+8. The model detail link points at an `ap-southeast-1` console entry; the notice leaves the launch scope open. - **ID and price:** `dreamina-seedance-2-5-flash-261115` has its own row in the pricing tables, billed per million tokens. - **Limits:** 480p and 720p only; 1080p, 4K and offline inference are marked "Not supported". - **Until then:** standard Seedance 2.5 is callable on BytePlus and on OmniaKey. ## What BytePlus has announced for November 15 ModelArk's Create video generation task page carries a model availability notice for Dreamina Seedance 2.5 flash. The notice dates the opening to 14:00 UTC+8 on November 15, 2026 and covers online experience and API calls. It also opens by saying Flash "is now available", and the [Seedance 2.5 tutorial](https://docs.byteplus.com/en/docs/modelark/seedance-2-5) repeats both sentences. The rest of the docs still describe the standard model. The [model list](https://docs.byteplus.com/en/docs/modelark/model-list) leaves Flash out, the release announcement covers standard 2.5 only, and the product updates page ends at October 2026. The [model detail page](https://ai.byteplus.com/ark/region:ap-southeast-1/model/detail?name=dreamina-seedance-2-5-flash) is the console entry BytePlus links for the model. Treat the clock time as the plan, and confirm account access in your own ModelArk account after it passes. The two sentences in the notice describe different states, so the account is the place where the question gets settled. |Confirmed by BytePlus|Not published in any page checked| |-|-| |Opening: Nov 15, 2026, 14:00 UTC+8|Maximum output duration| |Entry: ModelArk model-detail link (ap-southeast-1)|Launch scope: regional or global| |480p and 720p output|Supported task types| |Token rates and 5 s example prices|Reference, audio and frame-rate limits| ## The exact ID and what the price list shows The [pricing page](https://docs.byteplus.com/docs/en/modelark/model-pricing.md) is the one place that names the model next to a rate: `dreamina-seedance-2-5-flash-261115`. Prices there are USD per million tokens, and video billing multiplies that million-token rate by consumption measured in millions. |Model|480p|720p|1080p| |-|-|-|-| |Flash|$0.34|$0.76|Not supported| |Standard 2.5|$0.514|$1.156|$2.843| Those figures are the vendor's estimate for one 5-second, 16:9 clip with no video input, quoted per video. The rates behind them are $7.0 per million tokens at 480p or 720p input without video, and $4.3 with video input. Standard 2.5 lists $10.70 and $6.40 for the same two cases. A video input lowers the unit rate and adds billable tokens, so the lower rate alone does not determine the total. > The 5-second example is a billing sample, not a duration cap. BytePlus has published no maximum output length for Flash. The billing rule is the per-million-token rate × token consumption measured in millions. The duration, size and frame-rate formula estimates raw tokens: `(input video duration + output video duration) × width × height × frame rate / 1024`. The charged token count is the `usage.completion_tokens` value in the response, not that estimate, and only successfully generated videos are charged. At both published resolutions, Flash costs more than the cheapest tier. Seedance 2.0 Fast lists $0.60 for the same 5-second 720p clip against $0.76 for Flash, so a low-cost test pipeline still has a cheaper option today. ## Flash and standard Seedance 2.5: what differs The published differences are the ID, the resolution ceiling and the price. Standard Seedance 2.5 covers 480p, 720p and 1080p, priced at $0.514, $1.156 and $2.843 for the same 5-second clip. Flash stops at 720p. The rest of the standard model's published capabilities have no Flash counterpart. BytePlus credits standard Seedance 2.5 with up to 30 seconds in one generation, two extensions, up to 50 omni reference assets, native audio, 24 fps and a task list that runs from text-to-video through first-frame, first-last-frame, reference, editing and extension. None of those claims appear next to Flash in the model list or the parameter tables. Our [Seedance 2.5 review](https://omniakey.com/blog/seedance-2-5-review) documents those capabilities for the standard model. Speed is unpublished as well. None of the pages checked reports a speed figure for Flash, and the name carries none either. ## What to prepare before you switch 1. Confirm the account after the opening. Check the region the console entry opens in (`ap-southeast-1` in the link BytePlus gives), that the ModelArk project is active, and that the model ID is accepted at all. The notice's own contradiction is the reason to look rather than assume. 2. Wait for the Flash parameters. The Create video generation task reference's model and parameter tables list standard Seedance IDs only, and Flash's request shape and task types are unpublished. A standard-model workflow transfers only after those tables name Flash; confirm them first, then run a small same-prompt comparison before moving a pipeline. 3. Budget in tokens rather than clip prices. Log `usage.completion_tokens` per job and compare it with the estimate; the $0.34 and $0.76 rows are samples for one short clip. 4. Keep a 720p fallback ready. Build the shot on the standard model first, using the [Seedance 2.5 prompt guide](https://omniakey.com/blog/seedance-2-5-prompt-guide), whose six clips and prompts all come from standard Seedance 2.5, then repeat the same prompt on Flash and compare the two once access and parameters are confirmed. ## What you can use today Standard Seedance 2.5 is on sale on BytePlus and in the OmniaKey catalog as [Seedance 2.5](https://omniakey.com/dreamina-seedance-2-5-260628). The currently listed video models sit on [the AI video generator](https://omniakey.com/ai-video-generator), the full catalog on [the models page](https://omniakey.com/models), and the task shape for video work in the [video API docs](https://docs.omniakey.com/en/media/video). OmniaKey currently lists standard Seedance 2.5 and does not list Flash. |Situation|What to do before November 15| |-|-| |You need a finished clip this week|Use standard Seedance 2.5| |You need 1080p|Use standard Seedance 2.5| |You need audio or reference assets|Use standard Seedance 2.5| |You are budgeting a Flash pipeline|Price it in tokens, then re-check the ID after the opening| |You are choosing a 720p tier on price|Compare Seedance 2.0 Fast as well| ## Frequently asked questions ### Is Seedance 2.5 Flash available right now? BytePlus schedules online experience and API calls from 14:00 UTC+8 on November 15, 2026. The same notice also says the model is now available, so check your ModelArk account after that time before relying on either sentence. ### What is the Flash model ID? `dreamina-seedance-2-5-flash-261115`. The shorter `dreamina-seedance-2-5-flash` appears only in the model detail URL, as a page name. ### Is Flash on OmniaKey? No. The catalog and the public API docs list four Seedance models: standard 2.5, 2.0, 2.0 Fast and 2.0 Mini. If and when Flash is added, the model page and the video API docs will name it. ### Does Flash replace standard Seedance 2.5? Price and resolution alone are not enough to recommend a replacement. Standard 2.5 keeps 1080p and publishes duration, reference and audio capabilities; keep it for needs it already covers, and evaluate Flash on those tasks once access opens. ## Sources checked - BytePlus ModelArk, [Create a video generation task API](https://docs.byteplus.com/en/docs/modelark/create-video-generation-task-api): availability notice with the November 15, 2026 time, checked November 13, 2026 UTC. - BytePlus ModelArk, [Model pricing](https://docs.byteplus.com/docs/en/modelark/model-pricing.md): Flash ID, token rates, 5-second examples and billing rules, checked November 13, 2026 UTC. - BytePlus ModelArk, [Model list](https://docs.byteplus.com/en/docs/modelark/model-list): video model tables without a Flash row, checked November 13, 2026 UTC. - BytePlus ModelArk, [Dreamina Seedance 2.5 tutorial](https://docs.byteplus.com/en/docs/modelark/seedance-2-5): standard-model capabilities and the repeated notice, checked November 13, 2026 UTC. - BytePlus ModelArk, [Seedance 2.5 Flash model detail](https://ai.byteplus.com/ark/region:ap-southeast-1/model/detail?name=dreamina-seedance-2-5-flash): the console entry BytePlus links for the model. Updated November 13, 2026, 02:59 UTC. We will revise the guide when the BytePlus model list or its capability and parameter tables list `dreamina-seedance-2-5-flash-261115`.