Pipecat HeyGen LiveAvatar: duration by tier vs a 60 second Sume job
In Pipecat, HeyGen LiveAvatar session length depends on your subscription tier. On Sume the length rule is fixed: 4 to 60 seconds per job, priced per second.
What does Pipecat say about the HeyGen limit?
The Pipecat page covers HeyGen's LiveAvatar over WebRTC. It says the session duration ceiling is set by the LiveAvatar subscription tier. It lists quality options low, medium, high and very_high, and notes that VP8 is deprecated. It states no latency figures, so none are repeated here.
| Item | What the page says |
|---|---|
| Transport | WebRTC |
| Session length | Ceiling set by the LiveAvatar subscription tier |
| Quality options | low, medium, high, very_high |
| VP8 | Deprecated |
| Latency | No figures stated |
How does Sume set length?
By the job, not by the plan. Every Sume avatar video is 4 to 60 seconds. Price scales by second and by tier, so you can compute it before you submit: standard is $0.184 a second, plus is $0.245 and max is $0.55 with no product image.
A 45 second plus clip is 45 x $0.245 = $11.025. The same clip at standard is $8.28.
| Tier | No product image | With product image |
|---|---|---|
| standard | $0.184 | $0.194 |
| plus (default) | $0.245 | $0.258 |
| max | $0.55 | $0.58 |
Which should I use?
A subscription ceiling suits an interactive agent that talks for minutes. A per-job window suits a scripted spot. If you only need a recorded explainer, pick a tier and read the quote from the table. The comparison with the HeyGen demos is in the earlier post.
Sources
Related posts
More in Sume Avatar 1.0
- Pipecat Simli max_session_length vs Sume's 4 to 60 second clip
Simli in Pipecat caps a live session with max_session_length and max_idle_time. Sume has no session: an avatar job is a 4 to 60 second file.
- Pipecat TavusVideoService live avatar, or a recorded Sume clip?
Pipecat's Tavus service speaks your agent's TTS live over WebRTC. A Sume avatar job is a 4 to 60 second recorded MP4. How to pick, by what the viewer does.
- Reuse one AI presenter across lip-sync clips with an avatar handle
Create the presenter once, then pass avatar_handle to H3 Max lip sync for every clip. Same face each time, no re-upload, one Idempotency-Key per line.
- Tavus Griffin-Lite is invite-only: what can you build today?
Griffin-Lite is a closed Tavus preview. Until you get in, Sume ships rendered avatar clips, lip sync and motion control for talking-head video.
Written by Sume