← Models

Sora 2

OpenAI · Sora

OpenAI's current Sora generation for cinematic text/image-to-video creation in product and API workflows.

Part of Sora family · Other versions: Sora 2 Pro
Type
video
Context
N/A
Max Output
N/A
Status
current
API Access
Yes
License
proprietary
video-generation text-to-video image-to-video openai creative
Released September 2025 · Updated April 8, 2026

Overview

Freshness note: Model capabilities, limits, and availability can change quickly. This profile is a point-in-time snapshot last verified on April 8, 2026.

Sora 2 is OpenAI’s current base API tier in the Sora family for high-fidelity video generation from text and image inputs. It targets creative and product teams that need fast iteration on concept videos, short-form scenes, and previsualization assets while staying inside OpenAI’s broader multimodal stack.

Capabilities

Sora 2 is designed for prompt-driven video synthesis with synced audio and support for iterative creative direction. In practice, it is the route for teams that want many first passes quickly before deciding whether a shot deserves the slower, more expensive Sora 2 Pro path.

That makes it useful for motion ideation, campaign concept testing, rough-cut planning, storyboard validation, and any workflow where speed matters almost as much as fidelity.

Technical Details

Sora 2 is video-native, so token-based context and output limits are not the primary published constraints. In this repository, contextWindow and maxOutput are intentionally set to 0 and interpreted as N/A.

OpenAI’s current model docs position Sora 2 as the base synced-audio API route, with sora-2-pro above it for higher-fidelity generations and larger output dimensions.

Pricing & Access

OpenAI now publishes Sora Video API pricing directly on its API pricing page. The base sora-2 tier is listed at 0.10persecondfor720pportraitorlandscapeoutput.Higher−fidelityworknowclearlyroutesto‘sora−2−pro‘,whichcosts0.10 per second for 720p portrait or landscape output. Higher-fidelity work now clearly routes to `sora-2-pro`, which costs 0.30 per second at the same 720p sizes and $0.50 per second for the larger 1024x1792 / 1792x1024 format.

Product-surface quotas still vary, so teams should separate API budgeting from ChatGPT and Sora-app allowance assumptions.

Best Use Cases

Best for short-form concept videos, storyboard-to-motion iteration, ad creative previsualization, and teams that need a lower-cost OpenAI video route before escalating select shots to Sora 2 Pro.

Comparisons

Compared with Sora 2 Pro, Sora 2 is the faster and cheaper tier for everyday creation, while Pro is the better choice when synced audio quality and shot fidelity matter more than cost. Compared with Veo 3, tradeoffs typically center on ecosystem fit, motion style, and workflow controls. Compared with grok-imagine-video-1212, Sora 2 is often chosen when teams already rely on OpenAI routing for adjacent workloads.