Module: RunApi::Kling::Types

Defined in:: lib/runapi/kling/types.rb

Overview

Type definitions and constants for Kling video generation.

Defined Under Namespace

Classes: AiAvatarResponse, AsyncTaskResponse, CompletedAiAvatarResponse, CompletedImageToVideoResponse, CompletedMotionControlResponse, CompletedTextToVideoResponse, ImageToVideoResponse, MotionControlResponse, TextToVideoResponse, Video

Constant Summary collapse

TEXT_TO_VIDEO_MODELS = Text-to-video model variants. kling-3.0 supports multi-shot, sound, and elements; V2.x models support negative prompts and cfg_scale.

%w[
  kling-3.0
  kling-v2.5-turbo-text-to-video-pro
  kling-v2.1-master-text-to-video
].freeze

AI_AVATAR_MODELS = AI avatar lip-sync quality tiers, from highest to fastest.

%w[
  kling-ai-avatar-pro
  kling-ai-avatar-standard
  kling-ai-avatar-v1-pro
  kling-v1-avatar-standard
].freeze

IMAGE_TO_VIDEO_MODELS = Image-to-video model variants. V2.5 turbo and V2.1 pro support last-frame image control.

%w[
  kling-v2.5-turbo-image-to-video-pro
  kling-v2.1-pro
  kling-v2.1-standard
  kling-v2.1-master-image-to-video
].freeze

TEXT_TO_VIDEO_OUTPUT_RESOLUTIONS = Output resolution options. 4k is highest quality but slowest.

%w[720p 1080p 4k].freeze

MOTION_CONTROL_MODELS =

%w[kling-3.0].freeze

MOTION_CONTROL_OUTPUT_RESOLUTIONS =

%w[720p 1080p].freeze

MOTION_CONTROL_CHARACTER_ORIENTATIONS = Whether the character faces the direction from the video or the image.

%w[video image].freeze

MOTION_CONTROL_BACKGROUND_SOURCES = Whether the background comes from the reference video or the subject image.

%w[video image].freeze

ASPECT_RATIOS =

%w[16:9 9:16 1:1].freeze

DURATION_RANGE = Duration range for kling-3.0 (seconds).

(3..15)

MULTI_PROMPT_DURATION_RANGE = Per-shot duration range in multi-shot mode (seconds).

(1..12)

FIXED_DURATIONS = Fixed duration options for V2.x models (seconds).

[5, 10].freeze

MULTI_PROMPT_MAX_LENGTH = Maximum character length for each multi-shot prompt segment.