Is there a Runway Agent API? What the help pages say, and Sume's
The Runway pages I read describe Workflow Endpoints and an MCP, not an Agent API. Sume's Agent Completions takes a required spend cap and returns a run.

The Runway help pages I read describe Agent as an in-app chat experience and describe two programmable routes next to it: publishing a Workflow as an Endpoint, and connecting an MCP client. They do not describe an endpoint that sends Agent a prompt, so I would not assume one exists. Sume does ship that shape: POST /v1/agent/completions with model sume-agent.
What each page offers
| Route | Runway | Sume |
|---|---|---|
| Prompt to an agent by API | Not described on the pages I read | POST /v1/agent/completions |
| Saved pipeline by API | Workflow published as an Endpoint through the Developer Portal | POST /v1/formats/{handle}/{slug}/runs |
| Chat clients | MCP at https://mcp.runwayml.com/mcp | Hosted MCP at https://mcp.sume.com/mcp |
| Spend control | Developer credits for endpoints; account credits for MCP | generation_spend_cap_usd is required |
Sume's agent call
messages may hold system and user turns only, or you may send a single instruction. Attachments are images only today; other file types follow later. The response is a run: poll /v1/agent-runs for the result. There is no streaming and no thread continuation.
import json, os, urllib.request, uuid
body = {"model": "sume-agent",
"instruction": "Write a 3-line caption for a trail mug ad.",
"generation_spend_cap_usd": 2}
req = urllib.request.Request(
"https://api.sume.com/v1/agent/completions",
data=json.dumps(body).encode(), method="POST",
headers={"Authorization": "Bearer " + os.environ["SUME_API_KEY"],
"Content-Type": "application/json",
"Idempotency-Key": str(uuid.uuid4())})
with urllib.request.urlopen(req) as resp:
print(json.load(resp))How to decide
- Need a reusable, repeatable pipeline: a Format, on either platform.
- Need one open-ended request from code: Sume's Agent Completions.
- Need the vendor's chat inside your client: the MCP on each side.
Limits of this comparison
I read Runway's help pages, not its developer reference, so a feature that exists only there would not show up here. Check the Developer Portal for the current list before you build, and treat Sume's side as what its docs list today.
Sources
Related posts
More in Comparisons
- Kling 4.0 voice reference vs Sume reference_audio_urls
Kling 4.0 Omni Reference accepts voice references. Sume takes 1-3 reference_audio_urls on models such as MiniMax H3; here is what each source documents.
- Kling 4.0 vs Kling 3.0: the spec differences in one dated table
Length, resolution, HDR, references, keyframes, audio and prompt size, Kling 4.0 against 3.0 as stated on Kling's pages, plus how each maps to a Sume request.
- Kling V3 Pro vs Standard vs O3 per minute
Hedra's Artificial Analysis table puts Kling V3 Pro at $20.16 per minute and V3 Standard at $15.12. How to compare per-minute prices with Sume per-second SKUs.
- Live avatar agent or rendered video: D-ID vs Sume
D-ID V4 Expressive Visual Agents are live, LLM-connected avatars. Sume avatar video is rendered from a script in 4 to 60 s. Which one fits which job.
Written by Sume