GPT Image 2.5 is live — OpenAI's newest image model, targeted edits that leave the rest of the frame alone

Kling 4.0 Flash — Kuaishou's Fast Kling 4.0 Video Model

Kling 4.0 Flash is the fast tier of Kuaishou's Kling 4.0 lineup: text-to-video, image-to-video from a first frame, and Omni Reference, at 3–20 seconds and 720p. Flash opened to early-access users on September 28, 2026 and has no public API yet. It is coming to reAPI; Kling 3.0 Turbo is live today.

Coming soon — the API for this model is not available yet.

What is Kling 4.0 Flash?

Kling 4.0 Flash is the lighter, faster version of Kling 4.0, Kuaishou's fourth-generation AI video model. Kling's September 28, 2026 release note describes it as built for frequent creation, combining fast generation with reliable quality. The official Kling 4.0 Flash spec sheet lists text-to-video, image-to-video from a first frame, Omni Reference, 3–20 second clips, 720p output, and 8-bit SDR. Flash went to a limited early-access group on September 28, while the full Kling 4.0 is scheduled for October. As of October 1, 2026, Kling's API changelog has no Kling 4.0 Flash entry, so there is no public API for it yet.

Kling 4.0 Flash coming soon to reAPI

What you can build with this model

Real-world workflows and production use cases you can build and ship with this model.

Kling 4.0 Flash fast video drafts for iteration

Draft fast, render once

Kling built Kling 4.0 Flash for frequent creation. Use it to test a prompt, a camera move, or a pacing idea at 720p, then spend the bigger render on the take that works. It is the same draft-then-deliver loop people already run with Kling 3.0 Turbo, one generation newer.

Kling 4.0 Flash 20-second single video clip

Twenty seconds in one clip

Kling 4.0 Flash generates 3–20 second clips, five seconds more than the 15-second cap on Kling 3.0 and Kling 3.0 Turbo. A short ad, a product turn, or a four-beat scene fits in one Kling 4.0 Flash generation instead of two stitched clips.

Kling 4.0 Flash image-to-video with a first frame and references

Animate a still, on-model

Give Kling 4.0 Flash a first frame and it animates from that image, or add references through Omni Reference to keep a character or product on-model. Early testers report good motion and prompt understanding, with element consistency better than earlier Kling models but not perfect.

Why reAPI for Kling 4.0 Flash

Build on Kling 3.0 Turbo today

Kling 3.0 Turbo is live on reAPI now at /models/kling-3-0-turbo: fast text-to-video and image-to-video at 720p or 1080p. Ship on it today and switch to Kling 4.0 Flash when it can be called.

One key for every model

Your reAPI key already covers Kling, Seedance, Veo, Wan, and the rest. When Kling 4.0 Flash goes live, it is a new model id on the same /api/v1/videos/generations endpoint, not a new vendor account or a yearly plan.

Pay per second, no subscription

reAPI bills by the second of output, and credits never expire. Rates will be posted on this page at launch; Kuaishou has not published API pricing for Flash yet.

Kling 4.0 Flash vs Kling 3.0 Turbo

Both are the fast tier of their Kling generation. Kling 3.0 Turbo values are from the live model on reAPI; Kling 4.0 Flash values are from Kling's official September 28, 2026 spec sheet.

Capability
Kling 4.0 Flash
Kling 3.0 Turbo on reAPI
Clip length
3–20 s
3–15 s
Resolution
720p, 8-bit SDR
720p or 1080p
Inputs
Text, first frame, Omni Reference
Text, first frame
References
Omni Reference supported
First frame only
Model generation
Kling 4.0
Kling 3.0
Availability
Early access in Kling's app; no API yet
Live now

Kling 4.0 Flash values are from Kling's official release note and may change before an API ships. This table will be updated with API parameters at launch.

Get ready for Kling 4.0 Flash in three steps

  1. step 01

    Create an API key

    Sign up and create a key in the dashboard, no card required. The same key will work for Kling 4.0 Flash at launch.

    Open
  2. step 02

    Build on Kling 3.0 Turbo

    Wire up /api/v1/videos/generations with model kling-3-0-turbo today. The request and task flow will be the same for Kling 4.0 Flash.

    Open
  3. step 03

    Switch the model id

    When Kling 4.0 Flash goes live, change the model to kling-4-0-flash. Final parameters and per-second rates will be posted here on launch day.

    Open
docs/api/kling-4-0-flash

API reference

Drop-in code and the full parameter table.

curl -X POST https://reapi.ai/api/v1/videos/generations \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "kling-4-0-flash",
    "prompt": "A clay lighthouse keeper climbs a spiral staircase at night"
  }'

Parameter validation failed

Rejected synchronously with the failing field named. Check enum values, ranges, and URL fields — nothing is charged.

Authentication or balance issue

401 means a missing or invalid Bearer key; 402 means the reserve exceeds your balance. Manage keys and credits in the console.

Content or material rejected

Reference material and outputs pass automated moderation. Rejected tasks fail with a clear error and a full refund.

Task failed or timed out

A task that reaches a failed state is never charged — the reserve refunds automatically. Keep the task id and retry when ready.

Frequently asked questions

Common questions about this model.

Kling 4.0 Flash is the fast, lighter tier of Kling 4.0, Kuaishou's fourth-generation AI video model. Kling's official spec sheet lists text-to-video, image-to-video from a first frame, Omni Reference, 3–20 second clips, and 720p output in 8-bit SDR. Kling describes it as built for frequent creation, with fast generation and good value.

start building

Ready to ship?

Try it in the playground or grab an API key to integrate now.