Runway Gen-4.5
Runway Gen-4.5 is a state-of-the-art AI video model that excels in cinematic quality, motion realism, and prompt adherence for both text-to-video and image-to-video generation.
Overview
What is Runway Gen-4.5
Runway Gen-4.5 is a state-of-the-art AI video generation model released in December 2025, designed to produce cinematic, photorealistic video clips from text prompts or still images. It leads in motion quality, physical accuracy, and prompt adherence, supporting clip lengths from 2 to 10 seconds in multiple aspect ratios.
Running it privately on Venice
On Venice, Runway Gen-4.5 runs under an anonymized privacy tier — your prompts are never stored, profiled, or used for training. This ensures private, sovereign access to one of the world’s most advanced video models without Big Tech surveillance. You pay per clip, not a subscription, making high-quality generation accessible on demand.
Assessment
Strengths and limitations
- Ranked #1 on the Artificial Analysis Text-to-Video benchmark with 1,247 Elo points, surpassing Google Veo 3 and OpenAI Sora 2 Pro.
- Exceptional motion quality, physical accuracy, and temporal consistency in generated clips.
- Supports both text-to-video and image-to-video generation with precise prompt adherence.
- Capable of interpreting complex camera choreography, scene composition, and atmospheric details in prompts.
- Available on Venice with zero retention of prompts — private by design.
- No native audio generation: clips are silent.
- Limited to 10-second maximum clip length.
- Output resolution capped at 720p.
- Proprietary model: not open-source or self-hostable.
Samples
Sample outputs
Generated on Venice with our standard prompt suite — the same prompts we run through every model of this type, so you can judge it like-for-like.
Slow aerial drone shot gliding over a misty mountain valley at golden hour, sunlight piercing clouds onto a winding river, ancient pine forests on either side, ultra-smooth motion, professional color grading, atmospheric haze, 4K
Capabilities
What it supports
- Text to video
- Image to video
- Reference to video
- Native audio generation
Variants
Runway Gen-4.5 model variants
Runway Gen-4.5 runs on Venice as 2 variants of the same underlying model. Pick by what you're starting from: a written prompt, a still image, reference images, or an existing clip. Each variant is its own model id on the API; the generation quality is the same across the family.
| Variant | What it is | Clip lengths | Resolutions | Aspect ratios | Audio | Model ID |
|---|---|---|---|---|---|---|
| Text to Videoflagship | Generate a clip from a written prompt | 2s – 10s | — | 16:9, 9:16 | runway-gen4-5-text | |
| Image to Video | Animate a still image into motion | 2s – 10s | — | 16:9, 9:16, 1:1, 4:3, 3:4, 21:9 | runway-gen4-5 |
Capability data comes straight from the Venice model API and refreshes with every catalog ingest. The specs and pricing on this page are captured from the flagship variant; pass the model id of the variant you want to the API.
Runway Gen-4.5 Text to Video
Generate a clip from a written prompt. Supports clips of 2s – 10s, 16:9, 9:16 aspect ratios.
runway-gen4-5-textRunway Gen-4.5 Image to Video
Animate a still image into motion. Supports clips of 2s – 10s, 16:9, 9:16, 1:1, 4:3, 3:4, 21:9 aspect ratios.
runway-gen4-5Specifications
Datasheet
- Maker
- Runway
- Released
- December 1, 2025
- Modality
- Text-to-video, Image-to-video
- Clip lengths
- 2s – 10s
- Mode
- text-to-video
- Aspect ratios
- 16:9, 9:16
- Audio
- No
- Prompt limit
- 1,000 chars
- Privacy on Venice
- Anonymized — prompts not stored
- Available on Venice since
- Apr 2026
- License
- Proprietary
API
Call it from your code
Venice exposes this model through the REST API. Queue a generation with the model id.
curl https://api.venice.ai/api/v1/video/queue \
-H "Authorization: Bearer $VENICE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "runway-gen4-5-text",
"prompt": "Aerial drone shot over a misty mountain valley at golden hour"
}'
# Use the returned queue_id with https://api.venice.ai/api/v1/video/retrieve.
# Call /video/complete after downloading if needed.Pricing
What it costs on Venice
Pay per clip on Venice — price scales with resolution and duration (2s–10s), from $0.32.
New Venice accounts include a free daily allowance and 500 welcome credits — no credit card required.
Alternatives
How it compares
| Model | Max clip length | Strongest at | Open weights | Price (Venice) |
|---|---|---|---|---|
| Runway Gen-4.5 | 10s | Cinematic motion & physics | No | from $0.32 |
| Kling O3 Pro | — | Longer coherent scenes | No | from $0.46 |
| Vidu Q3 | — | High-resolution output | No | from $0.27 |
| Wan 2.7 Enhanced | — | Open-source flexibility | Yes | from $0.68 |
Benchmark leader in prompt adherence and motion realism.
Use cases
What it is good for
- 01Creating cinematic ad content with controlled camera movements and realistic physics.
- 02Animating still artwork or concept art into short video sequences using image-to-video.
- 03Generating storyboards and pre-visualizations for filmmakers.
- 04Producing social media content with high visual fidelity and prompt accuracy.
- 05Prototyping visual effects and motion design for VFX pipelines.
Prompting
Getting better results
Use explicit camera terms like 'dolly in', 'pan left', or 'aerial shot' for precise motion control.
Describe physics-based actions clearly — e.g., 'glass shatters with realistic momentum'.
For image-to-video, upload a high-contrast, well-lit image for best motion tracking.
Keep prompts under 1,000 characters and focus on one core action to avoid ambiguity.
Version history
First publicly available Runway video model.
Predecessor with strong image-to-video capabilities.
Current — enhanced text-to-video, top benchmark performance.
FAQ
Frequently asked questions
Runway Gen-4.5 is a state-of-the-art AI video generation model released in December 2025. It supports both text-to-video and image-to-video modes, producing cinematic, photorealistic clips up to 10 seconds long with high motion quality and prompt adherence.
On Venice, Runway Gen-4.5 starts at $0.32 per clip, with pricing scaling based on duration and resolution. A 2-second clip costs $0.32, and longer clips are priced proportionally.
No. Runway Gen-4.5 is a proprietary model developed by Runway and is not open-source. It is available commercially via API and platform access, including on Venice.
No. Runway Gen-4.5 generates video only — audio is not included in the output. Clips are silent by default.
Runway Gen-4.5 outputs video at 720p resolution. It supports multiple aspect ratios including 16:9, 9:16, 1:1, 4:3, 3:4, and 21:9.
Yes. The variant 'runway-gen4-5' supports image-to-video, allowing you to animate a still image with motion. You can upload an image and generate a 2–10 second clip based on a text prompt.
Runway Gen-4.5 ranks higher on the Artificial Analysis benchmark (1,247 vs 1,226 Elo) and excels in motion physics and prompt adherence. Google Veo 3 offers longer clips and native audio, but Runway leads in cinematic realism and controllability.
Yes. Venice runs Runway Gen-4.5 with anonymized privacy — your prompts are not stored, linked to your identity, or used for training. This ensures uncensored, private access to the model.
The maximum clip length is 10 seconds. You can generate clips of any duration from 2 to 10 seconds in increments of 1 second.
Run Runway Gen-4.5 privately
No prompt logging. No data used for training.