ZMIME
Video model comparison · 2 models · Updated Nov 2, 2026

Kling v2.5 Turbo vs Wan v2.5 Preview

Preferred for text to video: Too close to call · Preferred for image to video: Wan v2.5 Preview · Newest: Wan v2.5 Preview. Quality scores are blind human-preference votes; specs and prices come from each lab’s listings.

  1. Kling AI

    Kling v2.5 Turbo

    Released Sep 23, 2025

    • Text to video
    • Image to video
    • Arena1,276
    • From—
    • Max length—
  2. Alibaba (Qwen)

    Wan v2.5 Preview

    Released Sep 24, 2025Beta

    • Text to video
    • Audio input
    • Arena1,321
    • From—
    • Max length—
  3. Add a model

    Make it a three-way comparison.

01 — At a glance

Where they differ

Quality is the arena score from blind, side-by-side human votes. Price, length and resolution come from each lab’s own listings.

  • Preferred for text to videoToo close to callKling v2.5 Turbo 1,220 (#28) · Wan v2.5 Preview 1,245 (#25) — blind human votes
  • Preferred for image to videoWan v2.5 PreviewKling v2.5 Turbo 1,276 (#27) · Wan v2.5 Preview 1,321 (#22) — blind human votes
  • NewestWan v2.5 PreviewKling v2.5 Turbo Sep 23, 2025 · Wan v2.5 Preview Sep 24, 2025
  • Widest inputsSame inputsKling v2.5 Turbo: Text prompts, Reference images · Wan v2.5 Preview: Text prompts, Audio
  • Native audioNo native audioGenerates an audio track with the video: Kling v2.5 Turbo no · Wan v2.5 Preview no
  • Image to videoKling v2.5 TurboStarts from an uploaded image: Kling v2.5 Turbo yes · Wan v2.5 Preview no
  • Self-hostingNo open weightsBoth are available only through APIs and apps
  • Most API hostsSame reachKling v2.5 Turbo: 1 · Wan v2.5 Preview: 1
02 — Side by side

Every spec in one table

Dashes mean the data is not published.

Kling v2.5 Turbo vs Wan v2.5 Preview specifications side by side
SpecificationKling v2.5 TurboKling AIWan v2.5 PreviewAlibaba (Qwen)
Quality (blind human votes)
Text to video arenaElo-style score · higher is better1,220 ±17#28 of 48 · 2,111 votes1,245 ±9#25 of 48 · 36,378 votes (best)
Image to video arenaElo-style score · higher is better1,276 ±13#27 of 48 · 3,795 votes1,321 ±10#22 of 48 · 16,907 votes (best)
Output
ProducesVideoVideo
Inputs
Text promptsYesYes
Start image (image to video)YesNo
Video inputNoNo
Audio inputNoYes
Access
WeightsProprietaryProprietary
API hosts11
ReleasedSep 23, 2025Sep 24, 2025
03 — Questions

Which should you choose?

Which is better, Kling v2.5 Turbo or Wan v2.5 Preview?

For text to video, voters rate them about the same (Kling v2.5 Turbo 1,220 (#28) · Wan v2.5 Preview 1,245 (#25)). For image to video, voters prefer Wan v2.5 Preview (Kling v2.5 Turbo 1,276 (#27) · Wan v2.5 Preview 1,321 (#22)). Arena scores come from blind, side-by-side human votes on the same prompts; when the gap is inside both models’ error margins we call it a tie. Test your own prompts before you commit.

What is the difference between Kling v2.5 Turbo and Wan v2.5 Preview?

Kling v2.5 Turbo is a video model from Kling AI (Sep 23, 2025), taking text prompts and reference images and producing video. Wan v2.5 Preview is a video model from Alibaba (Qwen) (Sep 24, 2025), taking text prompts and audio and producing video.

Which is newer?

Kling v2.5 Turbo Sep 23, 2025; Wan v2.5 Preview Sep 24, 2025.

Which generates video with sound?

Generates an audio track with the video: Kling v2.5 Turbo no; Wan v2.5 Preview no.

Can I run any of these on my own hardware?

No open weights. Both are available only through APIs and apps.