| Resolution | 720p (1280x720, 720x1280, 960x960) on getimg.ai; Kling's release states no resolution |
|---|
| Clip length | 5 or 10 s (fal API); 5 s only (getimg.ai) |
|---|
| Frame rate | 24 fps (getimg.ai) |
|---|
| Input modes | Text-to-video and image-to-video (getimg.ai); first-frame control, no last frame |
|---|
| Image inputs | Up to 4 images for multi-image generation (fal API parameter) |
|---|
| Editing | Multimodal video editing: add, remove or replace content using images or text (Kuaishou release) |
|---|
| Aspect ratios | 16:9, 9:16, 1:1 (fal API) |
|---|
Resolution and clip length are disputed between hosts (720p vs reports of 1080p; 5 s vs 5 or 10 s), so the source is named in each cell. Release: https://macaubusiness.com/kling-ai-advances-to-the-2-0-era-empowering-everyone-to-tell-great-stories-with-ai/ ; fal: https://fal.ai/models/fal-ai/kling-video/v2/master/image-to-video · 来源
| Spec | Value |
|---|---|
| Master text-to-video, 5 s | $1.40 (USD) |
| Each additional second | $0.28 (USD) |
Price as listed on the fal.ai endpoint page, read 24 Oct 2026. · 来源
| Comparison | Result |
|---|---|
| Image-to-video vs Google Veo 2 | Win-loss ratio 182% |
| Image-to-video vs Runway Gen-4 | Win-loss ratio 178% |
Kuaishou's own internal GSB (Good-Same-Bad) evaluation, as stated at the launch event; not an independent benchmark. · 来源
Estimate from the version's published specs and reviews, 0-5; scores are only comparable within one criteria year. Scored as launched (Kling 2.0 Master, 15 April 2025). Resolution is disputed: a reseller spec table (getimg) and a glossary list 720p and 5 s for Master, while a developer page says 1080p, so 720p is used; duration 3 assumes the usual 5 or 10 second options (the same table gives 5 s only). Silent. Control 3: image-to-video from a first frame plus the new multi-elements editor (add, swap, delete content using images, text or a video clip); no audio or multi-image references. Speed 0: Decrypt measured 'up to 16 minutes for nearly 5 seconds' on online platforms (over the 300 s anchor). Realism 4: Decrypt's hands-on and Tom's Guide praise motion and detail, with high prompt sensitivity. macaubusiness.com
The new version marks a significant leap forward from Kling 1.6, offering enhanced prompt understanding, more fluid character movement, and improved visual aesthetics that users describe as looking "filmed, not generated."
features a fresh interface and significantly more accurate animation than its predecessor.
Kling videos through version 2 are often high in detail and consist of a lot of motion. It can be hit and miss depending on the complexity of your prompt.
读者上传的照片,与图钉带来源的媒体分开显示。你可以为吃到或看到的内容评分,在餐厅图钉上还能直接在照片上标记菜品。
第一项始终是图钉的来源。总体可信度是各份资料对上方所用开始和结束时间支持程度的加权平均;资料每比最新的一份旧 180 天,权重减半。
快手环球通讯社2025年4月15日发布的消息称,克林人工智能在北京主办了2.0模型的“从视觉到屏幕”发布会; Tom's Guide 和 Decrypt[2] 将发布热潮的日期定在同一周(2025 年 2 月 20 日的开发者页面列表未作解释且未使用)。
20 April 2025 hands-on across dynamism, illustration, text-to-video and coherence: fluid motion and a multi-elements editor, but very sensitive to small prompt changes and up to 16 minutes for about five seconds on some platforms.[1]
18 April 2025: calls Kling 2.0 the 'world's most powerful' model of its kind with 'significantly more accurate animation' than its predecessor, illustrated by five social-media clips.[1]
Reseller spec table lists 2.0 Master at 720p, 5 seconds, 24 fps with first-frame image-to-video control.[1]
Dates the global launch of Kling 2.0 Master to 15 April 2025, with prompt responsiveness and multimodal editing (replace, insert, remove).[1]
有遗漏或错误吗?用你自己的话说明:能佐证此图钉的链接、不同的开始或结束日期及理由,或缺失、有误的信息。AI 会对照此图钉的来源进行核实,搜索更好的来源,并添加任何支持你说法的页面。图钉自身的来源仍然最重要。AI 也会查看图片:显示的是别的东西或显示效果差的图片会被移到后面或替换。
原标题: Kling 2.0 Master Released
使用 Gemini 3.8 Flash 完成翻译。