Video models

Kling 4.0 Flash opens to everyone: 20-second video via API

Kling has made its fast, lower-cost Kling 4.0 Flash model available to all users on the web and the API, plus MCP, CLI and plugins for Photoshop, Premiere Pro and After Effects. It follows the 30-second Kling 4.0 launched on 15 October.

· Announced 4 November 2026 · By Heather S.

Scroll
An aerial shot of a ship's bow in a storm with a pianist playing a piano on deck, a sample from Kling 4.0's release notes
Image: Kling AI (Kling 4.0 release notes)

Kling AI announced on 4 November 2026 that Kling 4.0 Flash is “now available to everyone” on the Kling AI website and through the API. It can also be reached through MCP, CLI and plugins.

What Flash offers

Flash is the fast, lower-cost sibling of Kling 4.0, and Kling now describes it as “fully available” after a more limited release. According to Kling’s release notes and API update:

  • Native video from 3 to 20 seconds, at 720p and 8-bit SDR.
  • Text-to-video, image-to-video from a single first frame, and Omni reference-to-video, which combines images, videos and elements. A single task accepts up to 15 inputs.
  • Prompts up to 26,000 English characters, up to 10 reference images and up to 5 reference videos.
  • Aspect ratios including 21:9 and adaptive.
  • API paths /text-to-video/kling-4.0-flash, /image-to-video/kling-4.0-flash and /omni-video/kling-4.0-flash. Billing is by total duration, including any input video; Kling points to its pricing page rather than the update notice for rates.

Background: Kling 4.0

The full Kling 4.0 became available to everyone on 15 October. Kling’s specifications list clips of 3 to 30 seconds, 720p, 1080p and 4K, 10-bit HDR at 1080p and above, up to 10 keyframe images, two-channel stereo audio and many languages and accents. The API followed on 28 October with output up to 30 seconds; the API notice lists 720p and 1080p, with 4K “coming soon”. Kling’s notes also say MCP and CLI support Kling 4.0, including Omni Reference and keyframe video.

Limits

Flash is shorter and lower resolution than the full model. Kling’s published claims about quality and “excellent value” come from its own materials, and we have not seen independent tests.

Our take

The more interesting detail is the delivery route. A video model offered through MCP, CLI and plugins is something an agent can call as a step in a pipeline, not only something a person types into a web page. For teams building interactive stories, a cheap 20-second model is a sensible choice for previsualisation and drafts, with the larger model kept for final shots. We discuss how models are becoming components of games in AI models as game generators.

Sources

← All news