kling-v3.0-pro
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
ChineseKling Video v3.0 Pro
Kling Video v3.0 Pro
Generate high-end cinematic video with synchronized audio through Frevana's OpenRouter video endpoint using .
kwaivgi/kling-v3.0-pro通过 Frevana 的 OpenRouter 视频端点使用 生成带有同步音频的高端电影级视频。
kwaivgi/kling-v3.0-proModel Characteristics
模型特性
- Model ID:
kwaivgi/kling-v3.0-pro - Resolution: (fixed)
720p - Aspect Ratios: (default),
16:9,9:161:1 - Durations: to
3seconds (default:15)5 - Frame Control: Supports both and
first_frame(image-to-video)last_frame - Input References: Reference images (); audio and video references are unsupported
image_url - Audio: Synchronized audio generation supported (default: enabled)
- Seed: Deterministic seed is not supported by this model
- Passthrough Controls: ,
negative_promptcfg_scale
- Model ID:
kwaivgi/kling-v3.0-pro - Resolution:(固定)
720p - Aspect Ratios:(默认)、
16:9、9:161:1 - Durations:至
3秒(默认:15)5 - Frame Control:同时支持 和
first_frame(图生视频)last_frame - Input References:参考图像();不支持音频和视频参考
image_url - Audio:支持同步音频生成(默认:启用)
- Seed:此模型不支持确定性 seed
- Passthrough Controls:、
negative_promptcfg_scale
Prerequisites & Authentication
前置条件与身份验证
- (or
FREVANA_TOKEN) or--token(orFREVANA_API_KEY) in the environment--api-key - Optional (or
FREVANA_AGENT_APP_INSTANCE_ID) sent as--agent-app-instance-idx-frevana-agent-app-instance-id - ,
curl, andbashpython3
- 环境中需设置 (或
FREVANA_TOKEN)或--token(或FREVANA_API_KEY)--api-key - 可选设置 (或
FREVANA_AGENT_APP_INSTANCE_ID),并通过--agent-app-instance-id发送x-frevana-agent-app-instance-id - 、
curl和bashpython3
Invocation
调用方式
bash
undefinedbash
undefinedText-to-video: submit and wait for video completion
Text-to-video: submit and wait for video completion
bash skills/kling-v3.0-pro/scripts/generate_video.sh
--prompt "A cinematic close-up of an astronaut's visor reflecting a distant exploding nebula"
--duration 6
--aspect-ratio 16:9
--download-dir ./output
--prompt "A cinematic close-up of an astronaut's visor reflecting a distant exploding nebula"
--duration 6
--aspect-ratio 16:9
--download-dir ./output
bash skills/kling-v3.0-pro/scripts/generate_video.sh
--prompt "A cinematic close-up of an astronaut's visor reflecting a distant exploding nebula"
--duration 6
--aspect-ratio 16:9
--download-dir ./output
--prompt "A cinematic close-up of an astronaut's visor reflecting a distant exploding nebula"
--duration 6
--aspect-ratio 16:9
--download-dir ./output
Image-to-video with first and last frame
Image-to-video with first and last frame
bash skills/kling-v3.0-pro/scripts/generate_video.sh
--prompt "The camera smoothly circles the character in dramatic studio lighting"
--first-frame ./first.png
--last-frame ./last.png
--duration 5
--download-dir ./output
--prompt "The camera smoothly circles the character in dramatic studio lighting"
--first-frame ./first.png
--last-frame ./last.png
--duration 5
--download-dir ./output
bash skills/kling-v3.0-pro/scripts/generate_video.sh
--prompt "The camera smoothly circles the character in dramatic studio lighting"
--first-frame ./first.png
--last-frame ./last.png
--duration 5
--download-dir ./output
--prompt "The camera smoothly circles the character in dramatic studio lighting"
--first-frame ./first.png
--last-frame ./last.png
--duration 5
--download-dir ./output
Status check
Status check
bash skills/kling-v3.0-pro/scripts/generate_video.sh status --job-id <JOB_ID>
undefinedbash skills/kling-v3.0-pro/scripts/generate_video.sh status --job-id <JOB_ID>
undefinedAllowed Options
允许使用的选项
- : Text description of the video (required for create)
-p, --prompt TEXT - : Path to file containing prompt
--prompt-file PATH - : Model slug (default:
--model MODEL)kwaivgi/kling-v3.0-pro - : Duration in seconds (
--duration SEC-3, default:15)5 - :
--aspect-ratio RATIO,16:9,9:16(default:1:1)16:9 - :
--resolution RES(default:720p)720p - : Exact output dimensions; overrides
--size WIDTHxHEIGHTand--resolution--aspect-ratio - : Image path or URL for first frame
--first-frame, --image PATH/URL - : Image path or URL for last frame
--last-frame PATH/URL - : Add a reference image; repeat for multiple images
--reference-image PATH/URL - : Unsupported by this model
--reference-audio URL - : Unsupported by this model
--reference-video URL - : Generate synchronized audio (default: true)
--audio - : Disable audio generation
--no-audio - : Negative prompt (e.g. "blurry, distorted, artifacts")
--negative-prompt TEXT - : CFG guidance scale (e.g. 0.5)
--cfg-scale NUM - : Wait for job completion and download/output (default in CLI)
--wait - : Submit job and return immediately with Job ID
--no-wait - : Seconds between status checks (default:
--poll-interval SEC)10 - : Maximum seconds to wait before timeout (default:
--max-wait SEC)600 - : Directory to download completed MP4 video
--download-dir DIR - : Path to save output JSON or MP4 file
-o, --output FILE - : Output only the video URL or downloaded file path
-t, --text-only - : Print request payload JSON and exit without network calls
--dry-run - : Frevana Bearer token override
--token TOKEN - : Frevana API key override
--api-key KEY - : Agent App instance ID override
--agent-app-instance-id ID
- :视频的文本描述(创建任务时必填)
-p, --prompt TEXT - :包含 prompt 的文件路径
--prompt-file PATH - :模型 slug(默认:
--model MODEL)kwaivgi/kling-v3.0-pro - :时长,单位为秒(
--duration SEC-3,默认:15)5 - :
--aspect-ratio RATIO、16:9、9:16(默认:1:1)16:9 - :
--resolution RES(默认:720p)720p - :精确的输出尺寸;会覆盖
--size WIDTHxHEIGHT和--resolution--aspect-ratio - :首帧图像的路径或 URL
--first-frame, --image PATH/URL - :尾帧图像的路径或 URL
--last-frame PATH/URL - :添加参考图像;可重复使用以添加多张图像
--reference-image PATH/URL - :此模型不支持
--reference-audio URL - :此模型不支持
--reference-video URL - :生成同步音频(默认:true)
--audio - :禁用音频生成
--no-audio - :Negative prompt(例如 "blurry, distorted, artifacts")
--negative-prompt TEXT - :CFG guidance scale(例如 0.5)
--cfg-scale NUM - :等待任务完成并下载或输出结果(CLI 中默认启用)
--wait - :提交任务后立即返回 Job ID
--no-wait - :状态检查的间隔秒数(默认:
--poll-interval SEC)10 - :超时前等待的最长秒数(默认:
--max-wait SEC)600 - :下载已完成 MP4 视频的目录
--download-dir DIR - :保存输出 JSON 或 MP4 文件的路径
-o, --output FILE - :仅输出视频 URL 或已下载文件的路径
-t, --text-only - :打印请求 payload JSON 并退出,不发起网络调用
--dry-run - :覆盖 Frevana Bearer token
--token TOKEN - :覆盖 Frevana API key
--api-key KEY - :覆盖 Agent App instance ID
--agent-app-instance-id ID
Guardrails
约束与防护规则
- For semantic requests such as "720p landscape" or "square video", use with
--resolution.--aspect-ratio - Use only when the user explicitly requires exact
--sizepixels; it overrides and replaces both semantic dimension fields in the request payload.WIDTHxHEIGHT - Validate that is between 3 and 15 seconds.
--duration - Validate that is
--resolution.720p - Validate that is one of
--aspect-ratio,16:9, or9:16.1:1 - is format-checked locally but not matched against a static model allowlist; model/provider compatibility is validated by the API.
--size - Do not pass ; Kling Pro does not support deterministic seed.
--seed - Frame images and input references may be sent together; OpenRouter gives precedence and treats the request as image-to-video.
frameImages
- 对于“720p landscape”或“square video”等语义请求,使用 配合
--resolution。--aspect-ratio - 仅当用户明确要求精确的 像素时使用
WIDTHxHEIGHT;它会覆盖并替换请求 payload 中的两个语义尺寸字段。--size - 验证 是否处于 3 至 15 秒之间。
--duration - 验证 是否为
--resolution。720p - 验证 是否为
--aspect-ratio、16:9或9:16。1:1 - 仅在本地检查格式,但不会与静态模型允许列表进行匹配;模型或提供商兼容性由 API 验证。
--size - 不要传入 ;Kling Pro 不支持确定性 seed。
--seed - 帧图像和输入参考可以同时发送;OpenRouter 会优先使用 ,并将请求视为图生视频。
frameImages