Google Veo 3.1
Google's high-quality video model with native audio, reference-image support, and frame-to-frame control.
From Google, Veo 3.1 is a cloud video-generation model. It produces short video clips from a text prompt, a starting image, or by interpolating between a first and last frame. It can also take reference video to guide the result. Clips run up to 8s at 4K with a synchronized audio track. It also supports a negative prompt, seed control, and reference images. It runs through OpenRouter, Runware, and Replicate using your own API key, from about $0.40 per second of output.
- Released
- Oct 2025
- Pricing
- $0.40 per second
- Max resolution
- 4K
- Aspect ratios
- 16:9, 9:16
- Clip length
- Up to 8s
- Audio
- Synchronized track
- Image-to-video
- Start & end frame + reference images
- Negative prompt
- Supported
- Seed control
- Supported
- Text to video rank
- #13
- Image to video rank
- #14
Samples
Generated with Veo 3.1 via Runware. The same two prompts run against every video model in the catalog (one landscape, one portrait), so the only thing that changes between two models’ clips is the model.
Camera motion, scene coherence over time
Fluid motion, fine detail, temporal stability
Google and Google DeepMind build the Gemini family of multimodal models, the Imagen and Nano Banana image models, the Lyria music models, and the Veo video models.
deepmind.google ↗Google's flagship reasoning model for complex software engineering, deep mathematical and scientific problem solving and autonomous multi-step agents, with a shared 1M-token context.
Google's most capable Flash model, with clear gains over 3.7 Flash in software engineering, agentic tasks, and multi-step reasoning, plus vision and a 1M-token context window.
Google's most efficient multimodal Flash model: an updated reasoning stack with stronger coding and computer-use quality, at the same speed and scale as the rest of the Flash line.
Google's most capable multimodal model with deep reasoning and 1M token context.
Google's 26B mixture-of-experts model with long context and strong multimodal reasoning.
Google's flagship 31B dense model with 256K context and top-tier text and image understanding.
One dies in 70 days, one was outranked by its own maker, one still ships. The matchup, re-run for real: six clips, first takes only.
Forty-seven minutes. Three dollars and seventy-three cents. Every prompt, every cost, one API — and the finished ad, embedded.
OpenAI killed Sora. The Chinese open-weight stack took the top three on the arena. And a 10-second clip now starts at fifty cents.
2.88 million models on Hugging Face. 359 on the leaderboard. You can do almost any job with five. The case for curation over catalogs.
A 2026 field guide: what each modality is good at, what it costs, and three ways to chain them together.