Image: AI-generated editorial illustration
Runway Dev has published access and pricing details for Gemini Omni 1.2 Flash, the Google video model it says is available through its developer platform. The guide, published Nov. 9, 2026, says developers can generate clips billed per second by output resolution, with sound included in every result and no separate audio option.
Runway Dev published access and pricing details for Gemini Omni 1.2 Flash on Nov. 9, 2026, adding Google’s video model to its developer platform with per-second billing, resolution-based output tiers and support for text-to-video, image references and video editing.
The company says the model is Google’s official successor to Gemini Omni 1.1 Flash and that it became available on Runway Dev the same day Google launched the API. Runway Dev’s changelog says the model was available on the platform on Nov. 5, 2026, the date Google Cloud said Gemini Omni 1.2 Flash became generally available.
What does Runway Dev’s new API support?
Runway Dev says Gemini Omni 1.2 Flash is a multimodal video model that can generate clips from text, first-and-last frames, image and video references, and can also extend or edit existing clips. The model identifier in Runway Dev’s docs is gemini_omni_flash_1.2.
The documentation says the model supports output at 360p, 720p, 1080p and 4K, and that 360p is positioned as a low-cost draft size. It also says requests without an explicit size render at 720p landscape.
Runway Dev says clips return with native audio generated in the same pass, and that sound is included in every clip. There is no separate audio option or audio price. Requests that send audio return a 400 error, according to the guide.
The model accepts prompts up to 4,000 characters. Runway Dev also says the output length is 3 to 15 seconds, or auto.
How is Runway Dev charging for Gemini Omni 1.2 Flash?
Runway Dev lists pricing for Gemini Omni 1.2 Flash as 2 credits per second at 360p, 7 credits per second at 720p, 10 credits per second at 1080p and 21 credits per second at 4K. One Runway credit equals $0.01, which puts the listed rates at $0.02, $0.07, $0.10 and $0.21 per second, respectively.
The pricing page says the listed rates were current as of Nov. 9, 2026. It also says Gemini Omni 1.2 Flash defaults to 720p and has a 6-credit minimum per request.
Runway Dev says reference images or keyframes cost one additional credit each, or $0.01 per image, and input video costs one credit per second, or $0.01 per second. The company also says editing bills the input video’s duration at the output rate plus input-video and reference-image charges.
For extensions, the documentation says requests with duration:auto reserve 15 seconds up front, then settle to the generated duration rounded up to a whole second. The guide says extension requests are billed only for the footage added, while edit requests are billed on the length of the source clip. Failed tasks blocked by input moderation are still charged.
How can developers access the model?
Runway Dev says developers can use Gemini Omni 1.2 Flash through Playground, API and SDKs, a coding agent, Workflows and Model Router. API access requires a Runway Dev account, an API key, project credits, the RUNWAYML_API_SECRET environment variable, the model name and an X-Runway-Version header.
The docs provide a text-to-video endpoint and an example request using POST /v1/text_to_video with the 2024-11-06 API version header. Generation is asynchronous: the create request returns a task ID immediately and users must poll until the task reaches a terminal status. Runway Dev says THROTTLED means the task is queued, not rejected.
The company also says output URLs are temporary and should be downloaded when tasks succeed. It says one API key and one credit balance can be used across Runway’s own models and other labs’ models.
Runway Dev says no Google Cloud account is required to use the API, only a Runway Dev account and API key.
What changed when Google made the model generally available?
Google Cloud said Gemini Omni 1.2 Flash became generally available on Nov. 5, 2026 and described it as a model for business video use cases such as product videos, virtual try-ons and ads. Google also said the model’s 720p list price is $0.07 per second, 30% less than Gemini Omni 1.1 Flash, and that clip length increased to up to 15 seconds from 10 seconds.
Google Cloud said audio references are allowlisted, outputs carry a SynthID watermark and the model is backed by Google Cloud infrastructure with Provisioned Throughput. It also said business users can use the model in Gemini for business and Google Vids, while developers can access it via API.
Runway Dev says Gemini Omni 1.2 Flash has been available on its platform since the day Google’s API launched. For developers building video workflows, the practical takeaway is that the model is now exposed through Runway’s portal with explicit per-second pricing, multiple resolution tiers and support for generation, extension and editing in one API surface.
Sources
- Gemini Omni 1.2 Flash API: access, pricing and codedev.runwayml.com· Runway Dev
- API Pricing & Costsdocs.dev.runwayml.com· Runway Dev
- Available AI Modelsdocs.dev.runwayml.com· Runway Dev
- Gemini Omni 1.2 Flash now GA: cost-effective videocloud.google.com· Google Cloud
- API Changelog & Updatesdocs.dev.runwayml.com· Runway Dev


