Runway Characters session cap is 5 minutes; what Sume Avatar 1.0 caps
Runway Characters docs list a 5-minute session, 10,000-character personality and 2,000-character start script. Sume Avatar 1.0 renders 4-60 second clips.
Runway's Characters documentation lists a maximum of 5 minutes per session, a personality field of up to 10,000 characters and a start script of up to 2,000 characters. Sume Avatar 1.0 has none of these: it renders a finished 4 to 60 second talking clip from a script, with no live session, personality prompt or start script.
What Runway documents
From Runway's Characters core concepts, read 2026-10-02:
- Session credentials can be consumed only once.
- Recording URLs returned after a session are temporary, per the same page.
| Item | Documented value |
|---|---|
| Session length | Maximum 5 minutes |
| Personality | Max 10,000 characters |
| Start script | Max 2,000 characters |
| Session states | NOT_READY, READY, RUNNING, COMPLETED, FAILED, CANCELLED |
| Connection | WebRTC, real time |
What Sume's equivalents are
Sume's avatar video takes an avatar_handle and one of script or video_inputs. Total estimated length must land in 4 to 60 seconds. The result is a file you fetch from the job result, not a stream.
There is no personality field, because the avatar does not decide what to say. Whatever the clip says is the text you wrote.
Which problem each solves
A Runway Character answers questions live, up to 5 minutes at a time. A Sume avatar clip delivers a fixed message the same way every time, and you can review the first frame, regenerate stills and caption the final MP4.
For a support bot that must respond to a user, a live session is the right shape and Sume does not offer one. For a welcome video, a product explainer or a personalised message, a rendered clip is cheaper to review and easy to cache.
- Live conversation: Runway Characters.
- Fixed, reviewable message: Sume Avatar 1.0.
- Long session: Runway limits you to 5 minutes; Sume limits one job to 60 seconds.
Using both
Some teams might greet with a rendered clip and hand over to a live session. Sume does not integrate with Runway Characters, so that handover is your own page logic.
Sources
Related posts
More in Comparisons
- Segmind API vs Sume: PixelFlow workflows or saved Formats
Segmind turns visual PixelFlow graphs into API endpoints. Sume saves an agent thread as a Format you call over the API. How the two reuse recipes.
- Sonilo segment-level music controls vs Sume section markers
Sonilo's text-to-music lets you set styles and moods per section. On Sume you write section markers like [0:00-0:30] Intro: inside one 5000-character prompt.
- Sonilo video-to-music on fal.ai: 600 s of footage vs Sume's route
Sonilo's video-to-music model scores footage up to 600 seconds on fal.ai. Sume has no video-to-music call: inspect the clip, write a prompt, mix with Timeline.
- Soniox $0.10 per hour async vs Sume STT $0.01 per minute
Soniox lists async transcription at $0.10 an hour. Sume STT is $0.01 per audio minute, $0.60 an hour. Where the gap matters, and what Sume's job gives you.
Written by Sume