Model providers

Using Runway with OpenClaw

2 min read

Browse more in Model providers.

All model providers guides →

This guide shows you how to wire Runway into OpenClaw so your agents can generate videos using Runway’s hosted models. You will configure authentication, set Runway as the default video provider, and understand which Runway models back text-to-video, image-to-video, and video-to-video.

By the end, your OpenClaw agent will route video generation requests through Runway automatically.

Prerequisites

  • ✓An existing OpenClaw installation with access to the `openclaw` CLI.
  • ✓A Runway account with an API secret so you can set `RUNWAYML_API_SECRET` or `RUNWAY_API_KEY`.
  • ✓An agent in OpenClaw that you can talk to and ask to generate videos.

Steps

  1. 1

    Export your Runway API secret

    Set one of the supported environment variables so OpenClaw can authenticate to Runway. `RUNWAYML_API_SECRET` is the canonical option, but OpenClaw also accepts `RUNWAY_API_KEY`, and you only need to set one of them in your runtime environment.

    bash
    export RUNWAYML_API_SECRET="sk-runway-..."
  2. 2

    Onboard the Runway API key with OpenClaw

    Use the OpenClaw onboarding flow for Runway so the CLI knows which auth choice to use. This step ties your configured environment variable to the `runway` provider inside OpenClaw.

    bash
    openclaw onboard --auth-choice runway-api-key
  3. 3

    Set Runway as the default video provider

    Configure your agent defaults so video generation calls go to Runway without extra flags. 5`, which OpenClaw uses for text-to-video and image-to-video.

    bash
    openclaw config set agents.defaults.videoGenerationModel.primary "runway/gen4.5"
  4. 4

    Review the Runway video generation config block

    Confirm that your configuration reflects Runway as the primary video generation model. primary` key to be set.

    json
    {
  agents: {
    defaults: {
      videoGenerationModel: {
        primary: "runway/gen4.5",
      },
    },
  },
}
  5. 5

    Generate a video through your agent

    Talk to your agent and ask it to generate a video; OpenClaw routes the request to Runway automatically. 5` and exposes `16:9` and `9:16` aspect ratios, while video-to-video runs require the `runway/gen4_aleph` model.

Configuration

OptionDescriptionExample
RUNWAYML_API_SECRETCanonical environment variable that authenticates the bundled `runway` provider for hosted video generation.sk-runway-abc123
RUNWAY_API_KEYEnvironment variable alias that also authenticates the Runway provider; OpenClaw accepts this as an alternative to `RUNWAYML_API_SECRET`.sk-runway-alt-456
agents.defaults.videoGenerationModel.primarySets the default video generation model for agents; `runway/gen4.5` makes Runway the primary provider.runway/gen4.5

Troubleshooting

Video generation requests never complete and the agent appears to hang.

Runway uses a task-based API and OpenClaw polls `GET /v1/tasks/{id}` until the video is ready. Long-running generations can look like a hang, so wait for the polling to finish or check your Runway dashboard to confirm the task status.

Video-to-video prompts do not work while text-to-video works correctly.

Video-to-video currently requires `runway/gen4_aleph` specifically. 5`.

Frequently asked questions

Powered by Mem0

Add persistent memory to OpenClaw

Official Mem0 plugin for OpenClaw keeps context across chats and tools. Smaller prompts, lower cost, better continuity for your agents.

Read the official Mem0 guideOpen Mem0

More in Model providers

Model providers

Using Alibaba Model Studio with OpenClaw

Model providers

Using BytePlus with OpenClaw

Model providers

Using fal with OpenClaw

Model providers

Using GLM (Zhipu AI) with OpenClaw

Model providers

Using Hugging Face with OpenClaw

Model providers

Using Inferrs with OpenClaw