Runway Gen-4.5
Compare Gen-4.5 text-to-video, first-frame motion, API control, and professional finishing from Runway’s first research announcement to downstream production access.
- Posts archived
- Attachments examined
- Capabilities with evidence
What changed, when
Announcement dates in KST · Select a change or open its example.
Workflow threads
Compare workflows
Compare product availability and limits
Prompt to motion
- Runway app · · The paid rollout made text-to-video available in the app.
- Runway Dev · · Runway Dev brings the model to the text-to-video endpoint.
- ChatGPT · · Runway Creative lists Gen-4.5 inside ChatGPT.
Animate a first frame
- Runway app · · Paid users can supply a first frame alongside the prompt.
- Runway Dev · · The image-to-video API accepts a still image and supports two- to ten-second output.
- ChatGPT · No matching evidence in this archive.
Availability and integrations
- Runway app · · Runway says it is available on all paid plans.
- Runway Dev · · The API lists Gen-4.5 with two generation routes.
- ChatGPT · · The integration exposes Runway’s model lineup in ChatGPT.
Editorial delivery
Workflows over time
Announcement dates in KST. Earlier notices establish what was already available.
Update archive
Official titles with edited summaries and post-level sources.
Official examples and illustrations
Play video in its card or enlarge the example for its source and context.
Read the dated updates and sources
- · Gen-4.5 introduces a new text-to-video model
Runway previews direct text-to-video with improvements in motion, visual fidelity, and following detailed prompts. Its release uses examples of action, camera direction, and timing. Runway announced a rollout at release. Its notice described image-to-video, keyframes, and video-to-video as future controls.
- · Text-to-video opens to all paid plans
Runway completes its announced paid-plan rollout for Gen-4.5. First-frame image input is dated separately in the January changelog.
- · A still image can set the first frame
Gen-4.5 accepts a source image as the first frame alongside a prompt that describes the motion. The notice lists paid plans. The Turing Reel paper, published the following day, evaluates examples from this new route.
- · Runway Dev adds text-to-video and image-to-video
The API accepts text alone or an input image, with clip lengths from two to ten seconds. Runway exposes these two generation modes under its gen4.5 API identifier.
- · Tool Mode adds editorial exports
Gen-4.5 clips in Tool Mode can be exported as ProRes 4444 with PCM audio or as PNG image sequences. The app notice lists Pro+ plans. The export format is distinct from generation controls.
- · API output adds HDR and mastering formats
Gen-4.5 text-to-video and image-to-video add HDR10, HLG, 10-bit SDR, and high-bit-depth mastering choices including ProRes and OpenEXR. The API options differ from the app’s dated ProRes 4444 export feature.
- · Gen-4.5 HDR can deliver ACEScg OpenEXR sequences
Scene-referred ACEScg OpenEXR delivery adds separate options for ACES 1.3 and 2.0 pipelines. The API notice identifies Gen-4.5 HDR generations and the separate video-to-HDR endpoint.
- · Runway Creative brings Gen-4.5 into ChatGPT
Runway lists Gen-4.5 in the Runway Creative integration, where users can start generation work from ChatGPT. The announcement documents the integration and model lineup. Tool Mode export and API controls are documented separately.
Documentation and availability notes
- Runway research · Runway’s January 22 follow-up studies detection with paired real footage and Gen-4.5 clips generated from the same first frame. Its detection rate applies to that experiment.
- Runway Dev API · The current model guide lists separate text-to-video and image-to-video routes and links to Runway’s video output formats.