- 🎬 Text-to-video model from Runway, released April 2026.
- 🆕 Successor to the Gen-4 family of video models.
- 🎯 Runway reports improved motion quality, prompt adherence, visual fidelity.
- 👁️ Supports both text-to-video and image-to-video modes.
- 🔧 Accessible via Runway's developer API as the gen4.5 model.
- 🏢 Built by New York–based generative-AI company Runway.
- 📏 Generates high-definition clips from written prompts.
Runway AI, Inc. is an American generative AI company based in New York City, focused on research and commercial products for synthesizing video, images, and other multimedia. The lab is best known for its text-to-video and video-to-video generative models, which have helped…
Explore 1 more model by Runway (company) →Runway Gen-4.5 is the latest video generation model from Runway, the New York–based generative-AI company behind earlier publicly available video systems including the Gen-4 line. This entry represents the text-to-video pathway of Gen-4.5, while the model also supports image-to-video generation; both modes are exposed through Runway's developer API under the gen4.5 identifier. According to Runway, the model produces high-definition clips from prompts describing desired motion and action.
Compared with its same-family predecessors, Gen-4.5 shifts emphasis toward text-driven creation. Where the wider Gen-4 line centered on image-to-video workflows and world consistency, Gen-4.5 foregrounds prompt adherence so users can drive scenes directly from written descriptions. Runway states the new generation improves on motion quality, prompt adherence, and visual fidelity relative to prior Gen-4 releases.
Within Runway's current lineup, the companion Runway Gen-4.5 image-to-video variant shares this base model, while Runway Gen-4 Turbo and Runway Gen-4 Aleph remain distinct Gen-4 video offerings.
Runway introduced Gen-4.5 through its research site, where it positions the release as the next step in its video generation models. For independent benchmark comparisons, consult neutral evaluators directly, as the figures cited here are Runway's own reported characterizations rather than third-party measurements.
This About section is AI-generated from public sources via VeniceStats + Venice inference, with no human editing. It may contain inaccuracies.
Daily usage · Runway Gen-4.5 · Organic sellers
| Seller | Health | Rep↓ | Sales | Video | Routing | $ / s | $ / video |
|---|---|---|---|---|---|---|---|
| ▲ Apex Ant 0x73b4…e736 | — | 81.71 | — | — | #2 | $0.0729 | — |
| Venice.ai Proxy 0x1f22…18c9 | failingupstream error · 3.7 h | 86.84 | $97.53 | 19.5 min | #1 | $0.0891 | — |
"Best price" and the seller table are live Antseed catalog data (advertised $/1M tokens — or $ per generated image for unit-billed image models — not settled amounts). Reputation = buyer trust score (0-100, the Antseed SDK's own formula). "Routing" = the SDK's default buyer routing (what the VPR desktop app ships with): a trust ≥ 60 gate on the effective reputation, then cheapest-first among routable sellers; live failover state (per-peer cooldowns) is buyer-side runtime and not included. Model knowledge (TLDR, provider, About) via the VeniceStats enrichment layer. Advertised catalog, not the model used in any specific purchase. "Usage on Antseed" counts only settlements whose buyers share the per-model split on-chain (metadata v2/v3, opt-in), so every usage figure is a lower bound.