REDVIDEO
Log in
LIVE REDVIDEO CATALOG

AI VIDEO & IMAGE MODELS

Explore model families through the workflows and controls currently configured in RedVideo. Each guide combines live catalog facts, practical use cases, and a compatible path into the studio.

01 / model

Create video and images with the Wan model family

Explore Wan AI video and image workflows in RedVideo, from prompt and frame generation to reference-led motion, video extension, and image editing.

  • VIDEO WORKFLOWS
  • IMAGE CREATION
  • CHOOSE A VERSION
Explore page 02 / model

Create AI video with Wan 3.0

Create Wan 3.0 video from text, a frame pair, or mixed references — images, clips, and audio — with adjustable 3 to 30 second output in the RedVideo studio.

  • ONE ROUTE, FOUR WORKFLOWS
  • DURATION AND OUTPUT
  • ACCESS AND QUOTES
Explore page 03 / model

Explore Kling 3.0 creative workflows

Explore Kling 3.0 Standard, Fast, Omni, motion-control, reference-video, and image-generation workflows available in the RedVideo studio.

  • THREE VIDEO ROUTES
  • MOTION + STORY
  • KLING IMAGE
Explore page 04 / model

Create AI video with the Seedance 2.0 family

Create Seedance 2.0 video from text, images, first and last frames, or supported reference media with Standard, Fast, and Mini routes.

  • SHARED WORKFLOWS
  • OUTPUT CHOICES
  • ITERATE WITH INTENT
Explore page 05 / model

Create AI video with Seedance 2.5

Generate Seedance 2.5 video from text, images, frame pairs, or up to 50 reference assets — audio-only references included — with single-shot output up to 30 seconds.

  • BIGGER REFERENCE BUDGET
  • LONGER SINGLE SHOTS
  • CHOOSE 2.5 OR 2.0
Explore page 06 / model

Create video with Veo 3.1 Fast or Quality

Create fixed eight-second Veo 3.1 video with Fast or Quality routes for text, image, and supported frame or reference workflows in RedVideo.

  • TWO ROUTES
  • FRAME GUIDANCE
  • PROMPT + AUDIO
Explore page 07 / model

Generate and edit images with Nano Banana models

Generate and edit images with Nano Banana, Pro, 2, and 2 Lite routes, including reference-guided workflows and version-specific resolution controls.

  • ONE FAMILY, FOUR ROUTES
  • EDIT WITH REFERENCES
  • VERSION-SPECIFIC CONTROL
Explore page 08 / model

Create and edit with GPT Image models

Create and edit images with GPT Image 1.5 and GPT Image 2 in RedVideo using prompts, supported references, quality, ratio, and resolution controls.

  • TWO IMAGE ROUTES
  • REFERENCE-LED EDITING
  • COMPATIBLE OUTPUTS
Explore page 09 / model

Create and edit images with Seedream models

Generate and edit images with Seedream 4.5, Seedream 5.0, and Seedream 5.0 Pro using reference-guided workflows and version-specific output settings.

  • VERSION RANGE
  • PROMPT + REFERENCES
  • FROM FRAME TO MOTION
Explore page 10 / model

Create video with Grok Imagine models

Create Grok Imagine video from text or an image, or animate a frame with Grok Imagine 1.5 using supported duration, ratio, and managed-audio controls.

  • TWO ROUTES
  • OUTPUT CONTROLS
  • CREATIVE DIRECTION
Explore page
COMPLETE MODEL API CATALOG

EVERY CALLABLE MODEL API

Every model API you can call in RedVideo is listed here, including the variants already covered by the family guides above. Their names, descriptions, workflows, and Studio links come directly from the current product catalog, so each version stays individually selectable without creating thin pages for every variant.

ALL AI VIDEO MODELS

  • RedVideo Core V2image to video · first last frame · video extend

    RedVideo Core V2: next-generation house image-motion route with native audio, one-pass long clips, and per-second billing.

    Open model
  • RedVideo Coreimage to video · first last frame · video extend

    House image-motion route with managed audio.

    Open model
  • Wan 2.5image to video

    Original Wan image motion with three output sizes.

    Open model
  • Wan 2.6image to video

    Higher-detail Wan image motion up to 15 seconds.

    Open model
  • Wan 2.6 Flashimage to video

    Faster Wan image motion with optional generated audio.

    Open model
  • Wan 2.6 R2Vreference to video

    Reference-guided motion from up to five images or videos.

    Open model
  • Wan 2.7text to video · image to video · first last frame · reference to video · video extend · video edit

    Wan 2.7 route for text or a single starting frame.

    Open model
  • Wan 2.7 Spicyimage to video

    Spicy image motion with generated audio and reference-audio support.

    Open model
  • Wan 2.2 Spicyimage to video

    Spicy image motion at 480p or 720p without audio.

    Open model
  • Wan 3.0text to video · image to video · first last frame · reference to video

    Wan 3.0 for paid plans: up to 30s with image, video, and audio references.

    Open model
  • Wan 3.0 Primetext to video · image to video · first last frame · reference to video

    Wan 3.0 Prime: identical quality, much faster generation at a higher per-second price.

    Open model
  • Veo 3.1 Fasttext to video · image to video · first last frame · reference to video

    Fast Veo generation with a fixed 8-second output and text, frame, or reference-image guidance.

    Open model
  • Veo 3.1 Qualitytext to video · image to video · first last frame

    Higher-quality Veo generation with a fixed 8-second output from text or first and last frames.

    Open model
  • Kling 2.6text to video · image to video · motion control

    Standard or Pro generation with optional Pro audio.

    Open model
  • Kling 3.0 Standardtext to video · image to video · motion control

    Full Kling 3.0 quality ladder: Standard, Pro, or 4K.

    Open model
  • Kling 3.0 Fasttext to video · image to video

    Fast Kling 3.0 with second-by-second duration control.

    Open model
  • Kling 3.0 Omnireference to video

    Omni reference generation with automatic or manual storyboards.

    Open model
  • Seedance 1.5 Protext to video · image to video · first last frame

    Seedance Pro with camera lock and optional audio.

    Open model
  • Seedance 2.0 Standardtext to video · image to video · first last frame · reference to video

    Full-resolution Seedance 2.0, including native 4K.

    Open model
  • Seedance 2.0 Fasttext to video · image to video · first last frame · reference to video

    Faster Seedance 2.0 at 480p or 720p.

    Open model
  • Seedance 2.0 Minitext to video · image to video · first last frame · reference to video

    Lower-cost Seedance 2.0 at 480p or 720p.

    Open model
  • Seedance 2.5text to video · image to video · first last frame · reference to video

    Seedance 2.5 with 30s single-shot output and up to 50 reference assets.

    Open model
  • Grok Imaginetext to video · image to video

    Grok video generation from text or an image, with style control.

    Open model
  • Grok Imagine 1.5image to video

    Grok 1.5 image motion with automatic framing.

    Open model
  • Hailuo Standardimage to video

    Hailuo image motion with paired duration and resolution profiles.

    Open model
  • Hailuo Proimage to video

    Higher-tier Hailuo image motion with paired output profiles.

    Open model
  • MiniMax H3text to video · image to video · reference to video

    MiniMax H3 video generation from text, image frames, or mixed references with 2K output.

    Open model
  • HappyHorse 1.1text to video · image to video · reference to video

    Flexible text, image, and reference generation with second-by-second duration control.

    Open model
  • HappyHorse 1.0text to video · image to video · reference to video · video edit

    HappyHorse 1.0 with text, image, reference, and video-edit workflows.

    Open model

ALL AI IMAGE MODELS

  • RedVideo Imagetext to image · image edit

    Balanced in-house generation and editing with optional face enhancement.

    Open model
  • Z-Image Turbotext to image

    Fast prompt-to-image generation without reference images.

    Open model
  • Seedream 4.5text to image · image edit

    High-resolution generation and editing with optional references.

    Open model
  • Seedream 5.0text to image · image edit

    High-resolution generation and editing with optional references.

    Open model
  • Seedream 5.0 Protext to image · image edit

    High-fidelity generation and editing with up to ten references.

    Open model
  • Nano Bananatext to image · image edit

    Prompt generation and multi-reference editing with strong instruction following.

    Open model
  • Nano Banana Protext to image · image edit

    Prompt generation and multi-reference editing with strong instruction following.

    Open model
  • Nano Banana 2text to image · image edit

    Prompt generation and multi-reference editing with strong instruction following.

    Open model
  • Nano Banana 2 Litetext to image · image edit

    Fast 1K prompt generation or reference-guided editing.

    Open model
  • Flux 2 Protext to image · image edit

    Professional prompt generation and reference-guided composition with up to eight inputs.

    Open model
  • GPT Image 1.5text to image · image edit

    Detailed prompt generation guided by up to sixteen references.

    Open model
  • GPT Image 2text to image · image edit

    Latest GPT image model with nine ratio modes and up to 4K output.

    Open model
  • Wan 2.6 Imagetext to image · image edit

    Wan prompt generation guided by up to four references.

    Open model
  • Wan 2.7 Imagetext to image · image edit

    Fast Wan image generation guided by up to nine references.

    Open model
  • Wan 2.7 Image Protext to image · image edit

    Higher-detail Wan image generation guided by up to nine references.

    Open model
  • Qwen Image 2.0text to image · image edit

    Qwen generation with explicit output dimensions.

    Open model
  • Qwen Image 2.0 Protext to image · image edit

    Higher-fidelity Qwen generation with explicit output dimensions.

    Open model
  • Kling Image 3.0text to image · image edit

    Kling image generation with controlled output resolution and an optional reference.

    Open model