Veo 3.1

Overview

Veo 3.1 is Google’s video generation model, referred to in some updates simply as Veo, with notable improvements in portrait-mode video generation, stronger visual consistency, and higher-resolution upscaling to 1080p and 4K. It appears across Google’s AI ecosystem, including Gemini App and Flow by Google, and has also been surfaced through an OpenAI compatibility layer for drop-in API use.

For AI Product Managers, Veo 3.1 matters because it signals how fast generative video is becoming more product-ready: outputs are improving not just in creativity, but in practical dimensions like aspect-ratio support, character/object/background consistency, and deliverable resolution. These capabilities make Veo 3.1 more relevant for production workflows such as marketing content, creator tools, storytelling features, and multimodal product experiences where reliability and integration options matter as much as raw generation quality.

Key Developments

• 2026-01-14 — Google AI announced Veo 3.1 support for portrait mode, improved visual consistency across characters, objects, and backgrounds, and state-of-the-art upscaling to 1080p and 4K.
• 2026-01-15 — Demis Hassabis shared additional Veo 3.1 improvements, including enhanced expressiveness, portrait mode support, and 1080p/4K upscaling, with rollout in Gemini App and Flow by Google.
• 2026-03-21 — Philipp Schmid highlighted that Veo 3.1 became available as a drop-in model via the OpenAI compatibility layer, enabling video generation through the `/v1/videos` endpoint.

Relevance to AI PMs

• Evaluate product-fit for short-form and mobile-first video experiences. Portrait mode support makes Veo 3.1 more suitable for TikTok, Reels, Shorts, and in-app vertical content workflows, which is important when prioritizing consumer and creator use cases.
• Reduce quality risks in generative media features. Improved visual consistency across characters, objects, and backgrounds is directly relevant when designing user experiences that require continuity, such as scene-based storytelling, branded content, or guided creative tools.
• Speed integration and experimentation. Availability through Google surfaces like Gemini App and Flow by Google, plus a drop-in OpenAI-compatible API path, lowers switching costs for teams already prototyping with familiar endpoints and multimodal stacks.

Related

• Google / Google AI — The organization behind Veo 3.1 and the source of its product announcements and rollout.
• Gemini — Closely connected in Google’s multimodal ecosystem; Veo 3.1 appears alongside Gemini image capabilities in the OpenAI compatibility layer.
• Nano Banana — Mentioned as the Gemini image model paired with Veo 3.1 video support in compatibility-layer updates.
• OpenAI compatibility layer — Important integration path that makes Veo 3.1 accessible through familiar API patterns such as `/v1/videos`.
• Gemini App — One of the first user-facing Google products where Veo 3.1 improvements were rolled out.
• Flow by Google — Another Google surface receiving Veo 3.1 updates, suggesting workflow-oriented use cases for video creation.
• Demis Hassabis — Shared public updates about Veo 3.1’s enhanced expressiveness and rollout.
• Philipp Schmid — Highlighted Veo 3.1’s availability via the OpenAI compatibility layer.
• Portrait mode — A key feature update that expands Veo 3.1’s applicability to vertical-video products.
• Visual consistency — A core improvement area that makes outputs more reliable for real-world product experiences.