Skip to main content
Glama

Runway MCP Server by useapi.net (third-party)

Server Details

Runway video and images on your own plan: video-to-video, character animation, lip-sync, 4K upscale

Ownership verified
Status
Healthy
Uptime
100.0% over the last 24 hours
Last Tested
Transport
Streamable HTTP · MCP 2025-11-25
URL
ChatGPT
–
Claude
–

TDQS

A4.2/5.0

Scored across 25 tools

Disambiguation4/5

Most tools have clearly distinct resource+action purposes, and descriptions extensively cross-reference alternatives. However, the general create_video vs. three single-model Gen-4 routes, create_image vs. create_frames, and the two upscale tools overlap enough that an agent must read descriptions carefully to avoid misselection.

Naming Consistency5/5

All tools use the same runway_ prefix followed by a snake_case verb_noun pattern (create_video, get_task, list_assets, upload_asset, delete_asset). Model-specific suffixes like _gen4_5 and _gen4_turbo remain predictable and consistent with the convention.

Tool Count3/5

At 25 tools this is at the heavy end of the acceptable range and includes redundant legacy routes: create_gen4, create_gen4_5, create_gen4_turbo duplicate what create_video already covers, and create_frames overlaps with create_image. The set could be consolidated without losing capability.

Completeness4/5

The surface covers generation, asset upload/read/list/delete, task get/list/delete/cancel, account discovery, live credits, voices, and both general and specialized upscaling. Minor gaps such as video extension or more account administration are not present, but core workflows are fully supported.

Publisher details

Operator
useapi.net
Operator website
https://useapi.net
Vendor relationship
Independent
Privacy policy
Not provided
Restrictions
Paid useapi.net subscription ($15/month flat). Requires your own Runway account. · Publisher source

Available Tools

25 tools
runway_act_twoAnimate a character (Act-Two)AInspect

Animate a character with Runway Act-Two: a driving performance video (3-30 s) moves a character given as an image or video asset, with optional body control and expression intensity. For motion transfer with Kling use runway_create_video (model kling-3.0-motion-control); to change only the voice of a video use runway_act_two_voice. Waits up to ~50 s, then returns the video url and assetId, the failure reason or the taskId for runway_get_task. Uses credits, or none with exploreMode. POST /gen4/act-two.

Inputs: driving_assetId (required), character_assetId (required), seed, wait, +6 more.

runway_act_two_voiceSwap the voice of a videoAInspect

Replace the voice in a video (3-30 s) with one of Runway's voices (Act-Two voice swap), keeping the picture. Pick a voiceId with runway_list_lipsync_voices. To make a face speak new words use runway_lipsync; to animate a character use runway_act_two. Waits up to ~50 s, then returns the video url and assetId, the failure reason or the taskId for runway_get_task. Uses credits, or none with exploreMode. POST /gen4/act-two-voice.

Inputs: voiceId (required), video_assetId (required), wait, replyRef, +2 more.

runway_cancel_taskStop tracking a taskA
DestructiveIdempotent
Inspect

Stop this API tracking a running task (a taskId from runway_list_running_tasks) and free its maxJobs slot on the account. It does not cancel the task at Runway: the generation carries on and uses the account's credits, runway_get_task still reads it, but its replyUrl callback is no longer sent. 404 when the taskId is not being tracked. Cannot be undone. DELETE /scheduler/{taskId}.

Inputs: taskId (required).

runway_create_framesCreate Frames imagesAInspect

Generate 1 or 4 images at 1080p with Runway Frames from a required prompt, with up to 3 reference image assets cited as @IMG_1, @IMG_2, @IMG_3, a style preset and an aesthetic range (diversity). runway_create_image is the general image tool (13 models); runway_describe_image turns an image into a starting prompt for this one. Waits up to ~50 s, then returns each image's url and assetId plus the first image inline, the failure reason or the taskId for runway_get_task. Uses credits, or none with exploreMode. POST /frames/create.

Inputs: text_prompt (required), seed, wait, email, +11 more.

runway_create_gen4Create a Gen-4 videoAInspect

Generate a Runway Gen-4 video from a first-frame image asset (required) and an optional prompt, 5 or 10 s, through its own route; the account that holds the image runs it. runway_create_video with model gen4 does the same through the general route that covers every model (prefer it for new work); runway_create_gen4_turbo is the Gen-4 Turbo route. Waits up to ~50 s, then returns the video url and assetId, the failure reason or the taskId for runway_get_task. Uses credits, or none with exploreMode. POST /gen4/create.

Inputs: firstImage_assetId (required), seed, wait, seconds, +5 more.

runway_create_gen4_5Create a Gen-4.5 videoAInspect

Generate a Runway Gen-4.5 video through its own route: text-to-video (16:9) from text_prompt, or image-to-video from a first-frame image asset (firstImage_assetId) with an optional prompt, 5, 8 or 10 s. runway_create_video with model gen4.5 runs the same model with any length from 2 to 10 s and an audio switch, and covers every other model: prefer it for new work. Waits up to ~50 s, then returns the video url and assetId, the failure reason or the taskId for runway_get_task. Uses credits, or none with exploreMode. POST /gen4_5/create.

Inputs: seed, wait, email, maxJobs, +7 more.

runway_create_gen4_turboCreate a Gen-4 Turbo videoAInspect

Generate a Runway Gen-4 Turbo video from a first-frame image asset (required) and an optional prompt, 5 or 10 s, through its own route; the account that holds the image runs it. runway_create_video with model gen4-turbo does the same through the general route that covers every model (prefer it for new work); runway_create_gen4 is the standard Gen-4. Waits up to ~50 s, then returns the video url and assetId, the failure reason or the taskId for runway_get_task. Uses credits, or none with exploreMode. POST /gen4turbo/create.

Inputs: firstImage_assetId (required), seed, wait, seconds, +5 more.

runway_create_imageCreate images (all models)AInspect

Generate images with any of the 13 image models on Runway (FLUX 3, FLUX.2 Max / Klein, Nano Banana / 2 / 2 Lite / Pro, GPT Image 1.5 / 1 Mini / 2, Seedream 5, Gen-4, Gen-4 Turbo) from a prompt and up to 14 reference image assets, 1 or 4 images per task. The general image tool; runway_create_frames is the older Frames route (1080p, up to 3 references). Waits up to ~50 s, then returns each image's url and assetId plus the first image inline, the failure reason or the taskId for runway_get_task. Uses credits, or none with exploreMode (not flux-3). POST /images/create.

Inputs: model (required), seed, wait, email, +25 more.

runway_create_videoCreate a video (all models)AInspect

Generate a video with any of the 25 video models on Runway (Seedance 2.5 / 2.0, Wan 3.0, MiniMax H3, FLUX 3, Grok Imagine 1.5, HappyHorse, Kling O3 / 3.0 / 2.6, Veo 3.1, Gen-4.5, Gen-4, Gen-4 Turbo) from a prompt and optional reference assets (images, videos, audio from runway_upload_asset). The general video tool: runway_create_gen4_5, runway_create_gen4 and runway_create_gen4_turbo are older single-model routes; runway_edit_video_aleph, runway_act_two and runway_lipsync cover video-to-video, performance and lip sync. Waits up to ~50 s, then returns the video url and assetId (SUCCEEDED), the reason (FAILED) or the taskId for runway_get_task. Uses the account's credits, or none with exploreMode. POST /videos/create.

Inputs: model (required), seed, wait, audio, +79 more.

runway_delete_assetDelete an assetA
DestructiveIdempotent
Inspect

Delete one asset (an uploaded or generated file) from its Runway account by assetId; the account comes from the assetId. To remove a generation from the task history use runway_delete_task instead. Answers success: true; an unknown assetId answers 404. Cannot be undone. DELETE /assets/{assetId}.

Inputs: assetId (required).

runway_delete_taskDelete a taskA
DestructiveIdempotent
Inspect

Delete one task from its Runway account's task history by taskId; the account comes from the taskId. To delete a single file use runway_delete_asset; to stop this API tracking a running task (and free its slot) use runway_cancel_task. Answers success: true; an unknown taskId answers 404. Cannot be undone. DELETE /tasks/{taskId}.

Inputs: taskId (required).

runway_describe_imageDescribe an imageA
Read-onlyIdempotent
Inspect

Turn an image asset into a detailed text description (subject, scene, style) to edit and use as a prompt, for example in runway_create_frames or runway_create_image. Free and unlimited on every Runway account, the free plan too; answers at once, no task. Upload the image first with runway_upload_asset. GET /frames/describe/{image_assetId}.

Inputs: image_assetId (required).

runway_edit_video_alephEdit a video with AlephAInspect

Transform an existing video with Runway Gen-4 Aleph (video-to-video): a video asset plus a text prompt describing the change or style, optionally guided by an image asset. Use it to restyle or alter footage you have; to make a new video use runway_create_video, to edit with Kling O3 or Seedance 2.5 use runway_create_video (videoAssetId, video_mode). Waits up to ~50 s, then returns the video url and assetId, the failure reason or the taskId for runway_get_task. Uses credits, or none with exploreMode. POST /gen4/video.

Inputs: text_prompt (required), video_assetId (required), seed, wait, +4 more.

runway_get_assetGet an assetA
Read-onlyIdempotent
Inspect

One asset by assetId (from runway_upload_asset, runway_list_assets or a task's artifacts): name, mediaType, url, previewUrls, fileSize, media metadata (duration, width, height) and the taskId of a generated file. An unknown or deleted assetId answers 404. runway_list_assets lists many; runway_get_task reads a generation with its status. GET /assets/{assetId}.

Inputs: assetId (required).

runway_get_featuresGet credits and planA
Read-onlyIdempotent
Inspect

Ask Runway live for one account's plan credits (permitted.numPlanCredits) and how many are used (used.numPlanCredits), its storage allowance (permitted.storageGB) and the workspaces (teams) it belongs to with their currentPlan (not null = a paid workspace) and planExpiration; the rest of Runway's answer is passed through as is. Use before a generation to check credits; runway_list_accounts shows the stored settings without calling Runway. GET /features.

Inputs: email.

runway_get_taskGet a taskA
Read-onlyIdempotent
Inspect

One generation by taskId (from any runway_create_* tool or runway_list_tasks): status (PENDING, RUNNING, THROTTLED, SUCCEEDED, FAILED), progressRatio, options, the error of a failed task, and once SUCCEEDED the files in artifacts (url, assetId). wait: true waits up to ~50 s for an unfinished task; call again until it is SUCCEEDED or FAILED. An unknown or deleted taskId answers 404. GET /tasks/{taskId}.

Inputs: taskId (required), wait.

runway_lipsyncLip sync a faceAInspect

Make the face in an image or video asset speak: either an audio asset, or voice_text read by one of Runway's voices (voiceId from runway_list_lipsync_voices). Give exactly one of image_assetId / video_assetId and exactly one of audio_assetId / voiceId. To swap the voice of a whole video use runway_act_two_voice. Waits up to ~50 s, then returns the video url and assetId, the failure reason or the taskId for runway_get_task. Uses credits, or none with exploreMode. POST /lipsync/create.

Inputs: wait, maxJobs, voiceId, model_id, +7 more.

runway_list_accountsList Runway accountsA
Read-onlyIdempotent
Inspect

The Runway accounts added to this useapi.net API token, as stored: email, maxJobs (tasks at once), the paid workspace it bills against when set (id, name, plan), session issue and expiry times, and any stored account error; passwords and session tokens are never returned. email reads one account. No call to Runway: runway_get_features shows live credits and plan. Accounts are added on the setup page, not here. GET /accounts.

Inputs: email.

runway_list_assetsList assetsA
Read-onlyIdempotent
Inspect

List one Runway account's assets, uploads and generated files alike (web app included): assetId, name, mediaType, url, previewUrls, fileSize, media metadata (duration, width, height) and, for a generated file, the taskId that made it. Page with offset and limit (1-50, default 20); mediaType filters to video, image or audio. The list is scoped to the account's workspace while runway_get_asset reads by id unscoped, so a readable asset may be missing from the list. runway_get_asset reads one asset; runway_list_tasks lists the generations. GET /assets.

Inputs: email, limit, offset, mediaType.

runway_list_lipsync_voicesList lip sync voicesA
Read-onlyIdempotent
Inspect

List the voices Runway offers for lip sync and voice swap, live from the account: each voice's voiceId, name, labels (accent, age, gender, description, use case) and a sample MP3 url, all at once. Use it to pick the voiceId for runway_lipsync or runway_act_two_voice. GET /lipsync/voices.

Inputs: email.

runway_list_running_tasksList running tasksA
Read-onlyIdempotent
Inspect

The tasks this API is tracking now (started through it and not yet finished), each with its taskId and the replyUrl / replyRef given at create. The first check when a create tool answers 429: each account runs at most its maxJobs tasks at once (runway_list_accounts). Tasks started in the Runway web app are not here: use runway_list_tasks with statuses PENDING, RUNNING, THROTTLED. GET /scheduler.

Inputs: no named parameters.

runway_list_tasksList tasksA
Read-onlyIdempotent
Inspect

List one Runway account's task history, generations started in the Runway web app included: taskId, taskType, name (first 200 characters), status (PENDING, RUNNING, THROTTLED, SUCCEEDED, FAILED), createdAt, progressRatio, error and the finished files in artifacts (assetId, url, filename); runway_get_task gives a task's full record with its options. Page with offset and limit (1-1000, default 10); statuses filters, e.g. PENDING, RUNNING and THROTTLED for everything still running. runway_list_running_tasks shows only what this API is tracking now; runway_get_task reads one task. GET /tasks.

Inputs: email, limit, offset, statuses.

runway_upload_assetUpload an assetAInspect

Upload your own image, video or audio file to a Runway account's assets, for use as an input (an assetId) in runway_create_video, runway_create_image, runway_edit_video_aleph, runway_act_two, runway_lipsync and the other create tools; a generation that uses the asset runs on the account it was uploaded to. Send a public https url or contentBase64 (small files), up to 50 MB here; larger files go to POST /assets directly. Returns the new assetId, its url, name, type and size. POST /assets.

Inputs: name (required), url, email, width, +3 more.

runway_upscale_gen4Upscale a Gen-4 video to 4KAInspect

Upscale a Gen-4, Gen-4 Turbo, Gen-3 Alpha or Gen-3 Alpha Turbo video asset to 4K with Runway's own upscaler. For any other video (Kling, Veo, Seedance, Wan and the rest, or your own upload) use runway_upscale_video (Topaz). Waits up to ~50 s, then returns the video url and assetId, the failure reason or the taskId for runway_get_task. Uses credits, or none with exploreMode. POST /gen4/upscale.

Inputs: assetId (required), wait, replyRef, replyUrl, +1 more.

runway_upscale_videoUpscale a video to 4K (Topaz)AInspect

Upscale any video asset up to 40 s to 4K with Topaz, keeping its aspect ratio (a 1280x720 source becomes 4096x2304): a video from runway_create_video or any other Runway task, or your own upload. The account that holds the asset pays: 2 credits per second, or none with exploreMode on an Unlimited plan. runway_upscale_gen4 is the older 4K upscale for Gen-4 and Gen-3 Alpha (Turbo) videos only. Waits up to ~50 s, then returns the video url and assetId, the failure reason or the taskId for runway_get_task. POST /videos/upscale.

Inputs: videoAssetId (required), wait, maxJobs, replyRef, +2 more.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Related MCP Connectors

    Related MCP Servers

    Try in Browser

    Glama MCP Gateway

    Add one secure layer between your agents and this server.

    Resources