Veo 3.1

Overview

Veo 3.1 is Google’s updated video generation model, also referred to as Veo, with notable improvements in portrait mode support, visual consistency, and high-resolution upscaling to 1080p and 4K. Based on newsletter mentions, Veo 3.1 has been surfaced through Google’s ecosystem, including Gemini App and Flow by Google, and later exposed via an OpenAI compatibility layer for more standardized developer access.

For AI Product Managers, Veo 3.1 matters because it signals where generative video products are becoming more production-ready: better consistency across characters, objects, and backgrounds; support for portrait-oriented output suited to mobile and social channels; and higher-quality exports that fit real marketing, creator, and enterprise workflows. It also shows Google pushing multimodal generation into more interoperable deployment paths, lowering adoption friction for teams already building against familiar APIs.

Key Developments

• 2026-01-14 — Google AI announced a Veo 3.1 update for Ingredients to Video, adding 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 highlighted Veo 3.1 improvements including enhanced expressiveness, portrait mode support, and 1080p and 4K upscaling, with rollout in the Gemini App and Flow by Google.
• 2026-03-21 — Philipp Schmid shared that Veo 3.1 became available as a drop-in option via the OpenAI compatibility layer, enabling video generation through the `/v1/videos` endpoint.

Relevance to AI PMs

• Design for channel-specific video output: Portrait mode support makes Veo 3.1 directly relevant for products targeting mobile-first use cases such as short-form social video, in-app content generation, and creator tools.
• Improve generation reliability in user-facing workflows: Better visual consistency across scenes and subjects is important for PMs building templates, branded content flows, or multi-shot storytelling features where continuity affects user trust.
• Reduce integration friction for developer platforms: Availability through an OpenAI-compatible API suggests a practical path for teams to test Google video generation without reworking their entire tooling stack, which can accelerate experimentation and vendor comparison.

Related

• Google / Google AI — Veo 3.1 is part of Google’s generative media stack and was announced through Google AI channels.
• Gemini — Veo 3.1 is connected to the broader Gemini ecosystem, including multimodal product experiences and adjacent image model updates.
• Gemini App — One of the rollout surfaces for Veo 3.1 improvements, showing end-user product integration.
• Flow by Google — Another Google surface where Veo 3.1 updates were made available, relevant for creative and workflow-oriented use cases.
• OpenAI compatibility layer — Important for developers and PMs because it positions Veo 3.1 as easier to adopt through familiar API patterns like `/v1/videos`.
• Nano Banana — Mentioned alongside Gemini image model access in the same compatibility-layer update, indicating parallel multimodal rollout.
• Philipp Schmid — Publicly highlighted the compatibility-layer availability for Veo 3.1.
• Demis Hassabis — Announced key Veo 3.1 product improvements including expressiveness and portrait mode.
• Portrait mode — A defining capability update that expands Veo 3.1’s fit for mobile-native and social content products.
• Visual consistency — A core product improvement that makes generated video more usable in repeatable commercial workflows.