Agnes Video 2.5 Flash: Next-Gen Video, Free for a Limited Time

The newest generation of video generation: text-to-video, first/last-frame and image-reference modes, 4–12s durations, six aspect ratios — 720P at $0/sec for a limited time.

agnes-video-2.5-flashFree for a limited time

Agnes Video Generator has fully integrated Agnes AI's newest 2.5 series of video models. Video 2.5 Flash is the free edition (limited-time promotion); Video 2.5 is the paid flagship with higher quality (up to 2K). This page covers the capabilities, usage and known limits of Video 2.5 Flash — see the model comparison page for the differences between all three models.

Free for a limited time — $0/sec at 720P

During the promotion, generating with Video 2.5 Flash costs nothing (720P quality, per-second billing). No subscription or card required — add a free Agnes AI API key in the app and start generating.

Need more quality? The paid Video 2.5 adds 960P and 2K tiers ($0.025–$0.055/sec), supports reference videos, and includes the first 5 reference images free.

Three generation modes

Text to video

Type a prompt and generate a video directly — great for validating ideas fast.

First & last frame

Provide the first and/or last frame image; the model animates what happens in between — precise control over the start and end.

Image reference (up to 5 images)

Supply up to 5 reference images; the video keeps the characters, style and scene elements from them.

Key specifications

  • Durations of 4–12 seconds (whole seconds)
  • Six aspect ratios: 21:9 / 16:9 / 4:3 / 1:1 / 3:4 / 9:16
  • Fixed 720P output (e.g. 9:16 → 720×1280)
  • Up to 5 reference images
  • No negative prompt support (v2.0 does support it)

Known limits, explained

The 2.5 series uses a new submission protocol: output is defined by aspect ratio + quality tier rather than arbitrary pixels; duration is limited to 4–12s; negative prompts are not supported (Flash also lacks reference videos). If you need longer durations (up to ~17s), arbitrary pixel resolutions or negative prompts, the free legacy model Video 2.0 is still available. Agnes Video Generator adapts the form options to the selected model automatically — no manual parameter juggling required.

How to use it

  1. 1.Launch Agnes Video Generator (Web Studio, Docker or npm) and add your free Agnes AI API key.
  2. 2.Open the video model selector in the settings panel or the Simple Video form, compare the three models, then choose Video 2.5 Flash.
  3. 3.Enter your prompt / reference images, pick a duration and aspect ratio, and generate — the form only shows options the selected model supports.
Open the Web StudioTry the live demoCompare the three modelsv6.2.0 release notes

Related Pages

Why you can trust this

Written by a practicing developer and cross-checked against authoritative primary sources such as arXiv papers, vendor technical reports, and the Stanford AI Index.

S

SandGrid@lcy362

Author of Agnes Video Generator · Full-stack Developer

Independent developer and author of Agnes Video Generator, an open-source (MIT) AI video generation tool built on Agnes AI's free video models. Has helped 1,000+ creators produce AI videos at zero cost. Focused on making video generation models accessible and production-ready; content is based on hands-on practice and primary research.

View on GitHub

Last updated2026-08-21

Ready to Start Creating?

"Making world-class AI belong to everyone." — Bruce Yang. It's completely free, no credit card, and you won't need a high-end GPU. Your first AI video starts at zero cost. Want to use Agnes AI's free video models? This is the easiest way in.

Clone the GitHub repo and launch in 2 minutes